⬆️ Update traefik Docker tag to v3.6

* Sort host lists in state file for consistent output

Multi-host stacks (like glances) now have their host lists sorted
alphabetically when saving state. This makes git diffs cleaner by
avoiding spurious reordering changes.

.envrc.example

@@ -0,0 +1,6 @@
+# Run containers as current user (preserves file ownership on NFS mounts)
+# Copy this file to .envrc and run: direnv allow
+export CF_UID=$(id -u)
+export CF_GID=$(id -g)
+export CF_HOME=$HOME
+export CF_USER=$USER

.gitattributes

@@ -0,0 +1,2 @@
+*.gif filter=lfs diff=lfs merge=lfs -text
+*.webm filter=lfs diff=lfs merge=lfs -text

.github/check_readme_commands.py

@@ -0,0 +1,88 @@
+#!/usr/bin/env python3
+"""Check that all CLI commands are documented in the README."""
+
+from __future__ import annotations
+
+import re
+import sys
+from pathlib import Path
+from typing import TYPE_CHECKING
+
+if TYPE_CHECKING:
+    import typer
+
+from compose_farm.cli import app
+
+
+def get_all_commands(typer_app: typer.Typer, prefix: str = "cf") -> set[str]:
+    """Extract all command names from a Typer app, including nested subcommands."""
+    commands = set()
+
+    # Get registered commands (skip hidden ones like aliases)
+    for command in typer_app.registered_commands:
+        if command.hidden:
+            continue
+        name = command.name
+        if not name and command.callback:
+            name = getattr(command.callback, "__name__", None)
+        if name:
+            commands.add(f"{prefix} {name}")
+
+    # Get registered sub-apps (like 'config')
+    for group in typer_app.registered_groups:
+        sub_app = group.typer_instance
+        sub_name = group.name
+        if sub_app and sub_name:
+            commands.add(f"{prefix} {sub_name}")
+            # Don't recurse into subcommands - we only document the top-level subcommand
+
+    return commands
+
+
+def get_documented_commands(readme_path: Path) -> set[str]:
+    """Extract commands documented in README from help output sections."""
+    content = readme_path.read_text()
+
+    # Match patterns like: <code>cf command --help</code>
+    pattern = r"<code>(cf\s+[\w-]+)\s+--help</code>"
+    matches = re.findall(pattern, content)
+
+    return set(matches)
+
+
+def main() -> int:
+    """Check that all CLI commands are documented in the README."""
+    readme_path = Path(__file__).parent.parent / "README.md"
+
+    if not readme_path.exists():
+        print(f"ERROR: README.md not found at {readme_path}")
+        return 1
+
+    cli_commands = get_all_commands(app)
+    documented_commands = get_documented_commands(readme_path)
+
+    # Also check for the main 'cf' help
+    if "<code>cf --help</code>" in readme_path.read_text():
+        documented_commands.add("cf")
+    cli_commands.add("cf")
+
+    missing = cli_commands - documented_commands
+    extra = documented_commands - cli_commands
+
+    if missing or extra:
+        if missing:
+            print("ERROR: Commands missing from README --help documentation:")
+            for cmd in sorted(missing):
+                print(f"  - {cmd}")
+        if extra:
+            print("WARNING: Commands documented but not in CLI:")
+            for cmd in sorted(extra):
+                print(f"  - {cmd}")
+        return 1
+
+    print(f"✓ All {len(cli_commands)} commands documented in README")
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

.github/workflows/ci.yml

@@ -12,14 +12,14 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        os: [ubuntu-latest, macos-latest, windows-latest]
+        os: [ubuntu-latest, macos-latest]
         python-version: ["3.11", "3.12", "3.13"]
 
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6
 
       - name: Install uv
-        uses: astral-sh/setup-uv@v6
+        uses: astral-sh/setup-uv@v7
 
       - name: Set up Python ${{ matrix.python-version }}
         run: uv python install ${{ matrix.python-version }}
@@ -27,8 +27,8 @@ jobs:
       - name: Install dependencies
         run: uv sync --all-extras --dev
 
-      - name: Run tests
-        run: uv run pytest
+      - name: Run tests (excluding browser tests)
+        run: uv run pytest -m "not browser"
 
       - name: Upload coverage reports to Codecov
         if: matrix.os == 'ubuntu-latest' && matrix.python-version == '3.13'
@@ -36,13 +36,33 @@ jobs:
         env:
           CODECOV_TOKEN: ${{ secrets.CODECOV_TOKEN }}
 
+  browser-tests:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v6
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v7
+
+      - name: Set up Python
+        run: uv python install 3.13
+
+      - name: Install dependencies
+        run: uv sync --all-extras --dev
+
+      - name: Install Playwright browsers
+        run: uv run playwright install chromium --with-deps
+
+      - name: Run browser tests
+        run: uv run pytest -m browser -n auto -v
+
   lint:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6
 
       - name: Install uv
-        uses: astral-sh/setup-uv@v6
+        uses: astral-sh/setup-uv@v7
 
       - name: Set up Python
         run: uv python install 3.12
@@ -50,11 +70,5 @@ jobs:
       - name: Install dependencies
         run: uv sync --all-extras --dev
 
-      - name: Run ruff check
-        run: uv run ruff check .
-
-      - name: Run ruff format check
-        run: uv run ruff format --check .
-
-      - name: Run mypy
-        run: uv run mypy src/compose_farm
+      - name: Run pre-commit (via prek)
+        uses: j178/prek-action@v1

.github/workflows/docker.yml

@@ -0,0 +1,111 @@
+name: Build and Push Docker Image
+
+on:
+  workflow_run:
+    workflows: ["Upload Python Package"]
+    types: [completed]
+  workflow_dispatch:
+    inputs:
+      version:
+        description: 'Version to build (leave empty for latest)'
+        required: false
+
+env:
+  REGISTRY: ghcr.io
+  IMAGE_NAME: ${{ github.repository }}
+
+jobs:
+  build-and-push:
+    runs-on: ubuntu-latest
+    # Only run if PyPI upload succeeded (or manual dispatch)
+    if: ${{ github.event_name == 'workflow_dispatch' || github.event.workflow_run.conclusion == 'success' }}
+    permissions:
+      contents: read
+      packages: write
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v6
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Log in to Container Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ env.REGISTRY }}
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Extract version
+        id: version
+        run: |
+          if [ "${{ github.event_name }}" = "workflow_run" ]; then
+            # Get version from the tag that triggered the release
+            VERSION="${{ github.event.workflow_run.head_branch }}"
+            # Strip 'v' prefix if present
+            VERSION="${VERSION#v}"
+          elif [ -n "${{ github.event.inputs.version }}" ]; then
+            VERSION="${{ github.event.inputs.version }}"
+          else
+            VERSION=""
+          fi
+          echo "version=$VERSION" >> $GITHUB_OUTPUT
+
+      - name: Wait for PyPI
+        if: steps.version.outputs.version != ''
+        run: |
+          VERSION="${{ steps.version.outputs.version }}"
+          echo "Waiting for compose-farm==$VERSION on PyPI..."
+          for i in {1..30}; do
+            if curl -sf "https://pypi.org/pypi/compose-farm/$VERSION/json" > /dev/null; then
+              echo "✓ Version $VERSION available on PyPI"
+              exit 0
+            fi
+            echo "Attempt $i: not yet available, waiting 10s..."
+            sleep 10
+          done
+          echo "✗ Timeout waiting for PyPI"
+          exit 1
+
+      - name: Check if latest release
+        id: latest
+        run: |
+          VERSION="${{ steps.version.outputs.version }}"
+          # Get latest release tag from GitHub (strip 'v' prefix)
+          LATEST=$(gh release view --json tagName -q '.tagName' | sed 's/^v//')
+          echo "Building version: $VERSION"
+          echo "Latest release: $LATEST"
+          if [ "$VERSION" = "$LATEST" ]; then
+            echo "is_latest=true" >> $GITHUB_OUTPUT
+            echo "✓ This is the latest release, will tag as :latest"
+          else
+            echo "is_latest=false" >> $GITHUB_OUTPUT
+            echo "⚠ This is NOT the latest release, skipping :latest tag"
+          fi
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Extract metadata
+        id: meta
+        uses: docker/metadata-action@v5
+        with:
+          images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
+          # Only tag as 'latest' if this is the latest release (prevents re-runs of old releases from overwriting)
+          tags: |
+            type=semver,pattern={{version}},value=v${{ steps.version.outputs.version }}
+            type=semver,pattern={{major}}.{{minor}},value=v${{ steps.version.outputs.version }}
+            type=semver,pattern={{major}},value=v${{ steps.version.outputs.version }}
+            type=raw,value=latest,enable=${{ steps.latest.outputs.is_latest }}
+
+      - name: Build and push
+        uses: docker/build-push-action@v6
+        with:
+          context: .
+          push: true
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}
+          build-args: |
+            VERSION=${{ steps.version.outputs.version }}
+          cache-from: type=gha
+          cache-to: type=gha,mode=max

.github/workflows/docs.yml

@@ -0,0 +1,66 @@
+name: Docs
+
+on:
+  push:
+    branches: [main]
+    paths:
+      - "docs/**"
+      - "zensical.toml"
+      - ".github/workflows/docs.yml"
+  pull_request:
+    paths:
+      - "docs/**"
+      - "zensical.toml"
+      - ".github/workflows/docs.yml"
+  workflow_dispatch:
+
+permissions:
+  contents: read
+  pages: write
+  id-token: write
+
+concurrency:
+  group: "pages-${{ github.ref }}"
+  cancel-in-progress: true
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v6
+        with:
+          lfs: true
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v4
+
+      - name: Set up Python
+        run: uv python install 3.12
+
+      - name: Install Zensical
+        run: uv tool install zensical
+
+      - name: Build docs
+        run: zensical build
+
+      - name: Setup Pages
+        if: github.event_name != 'pull_request'
+        uses: actions/configure-pages@v5
+
+      - name: Upload artifact
+        if: github.event_name != 'pull_request'
+        uses: actions/upload-pages-artifact@v4
+        with:
+          path: "./site"
+
+  deploy:
+    if: github.event_name != 'pull_request'
+    environment:
+      name: github-pages
+      url: ${{ steps.deployment.outputs.page_url }}
+    runs-on: ubuntu-latest
+    needs: build
+    steps:
+      - name: Deploy to GitHub Pages
+        id: deployment
+        uses: actions/deploy-pages@v4

.github/workflows/release.yml

@@ -13,9 +13,9 @@ jobs:
     permissions:
       id-token: write
     steps:
-      - uses: actions/checkout@v4
+      - uses: actions/checkout@v6
       - name: Install uv
-        uses: astral-sh/setup-uv@v6
+        uses: astral-sh/setup-uv@v7
       - name: Build
         run: uv build
       - name: Publish package distributions to PyPI

.github/workflows/update-readme.yml

@@ -11,22 +11,24 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Check out repository
-        uses: actions/checkout@v4
+        uses: actions/checkout@v6
         with:
           persist-credentials: false
           fetch-depth: 0
 
       - name: Set up Python
-        uses: actions/setup-python@v5
+        uses: actions/setup-python@v6
 
       - name: Install uv
-        uses: astral-sh/setup-uv@v6
+        uses: astral-sh/setup-uv@v7
 
       - name: Run markdown-code-runner
         env:
           TERM: dumb
           NO_COLOR: 1
-          TERMINAL_WIDTH: 90
+          COLUMNS: 90  # POSIX terminal width for Rich
+          TERMINAL_WIDTH: 90  # Typer MAX_WIDTH for help panels
+          _TYPER_FORCE_DISABLE_TERMINAL: 1  # Prevent Typer forcing terminal mode in CI
         run: |
           uvx --with . markdown-code-runner README.md
           sed -i 's/[[:space:]]*$//' README.md

.gitignore

@@ -37,8 +37,13 @@ ENV/
 .coverage
 .pytest_cache/
 htmlcov/
+.code/
 
 # Local config (don't commit real configs)
 compose-farm.yaml
 !examples/compose-farm.yaml
 coverage.xml
+.env
+homepage/
+site/
+.playwright-mcp/

.pre-commit-config.yaml

@@ -1,4 +1,13 @@
 repos:
+  - repo: local
+    hooks:
+      - id: check-readme-commands
+        name: Check README documents all CLI commands
+        entry: uv run python .github/check_readme_commands.py
+        language: system
+        files: ^(README\.md|src/compose_farm/cli/.*)$
+        pass_filenames: false
+
   - repo: https://github.com/pre-commit/pre-commit-hooks
     rev: v5.0.0
     hooks:
@@ -10,18 +19,24 @@ repos:
       - id: debug-statements
 
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.8.4
+    rev: v0.14.9
     hooks:
-      - id: ruff
+      - id: ruff-check
         args: [--fix]
       - id: ruff-format
 
-  - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.14.0
+  - repo: local
     hooks:
       - id: mypy
-        additional_dependencies:
-          - pydantic>=2.0.0
-          - typer>=0.9.0
-          - asyncssh>=2.14.0
-          - types-PyYAML
+        name: mypy (type checker)
+        entry: uv run mypy src tests
+        language: system
+        types: [python]
+        pass_filenames: false
+
+      - id: ty
+        name: ty (type checker)
+        entry: uv run ty check
+        language: system
+        types: [python]
+        pass_filenames: false

.prompts/docs-review.md

@@ -0,0 +1,119 @@
+Review documentation for accuracy, completeness, and consistency. Focus on things that require judgment—automated checks handle the rest.
+
+## What's Already Automated
+
+Don't waste time on these—CI and pre-commit hooks handle them:
+
+- **README help output**: `markdown-code-runner` regenerates `cf --help` blocks in CI
+- **README command table**: Pre-commit hook verifies commands are listed
+- **Linting/formatting**: Handled by pre-commit
+
+## What This Review Is For
+
+Focus on things that require judgment:
+
+1. **Accuracy**: Does the documentation match what the code actually does?
+2. **Completeness**: Are there undocumented features, options, or behaviors?
+3. **Clarity**: Would a new user understand this? Are examples realistic?
+4. **Consistency**: Do different docs contradict each other?
+5. **Freshness**: Has the code changed in ways the docs don't reflect?
+
+## Review Process
+
+### 1. Check Recent Changes
+
+```bash
+# What changed recently that might need doc updates?
+git log --oneline -20 | grep -iE "feat|fix|add|remove|change|option"
+
+# What code files changed?
+git diff --name-only HEAD~20 | grep "\.py$"
+```
+
+Look for new features, changed defaults, renamed options, or removed functionality.
+
+### 2. Verify docs/commands.md Options Tables
+
+The README auto-updates help output, but `docs/commands.md` has **manually maintained options tables**. These can drift.
+
+For each command's options table, compare against `cf <command> --help`:
+- Are all options listed?
+- Are short flags correct?
+- Are defaults accurate?
+- Are descriptions accurate?
+
+**Pay special attention to subcommands** (`cf config *`, `cf ssh *`)—these have their own options that are easy to miss.
+
+### 3. Verify docs/configuration.md
+
+Compare against Pydantic models in the source:
+
+```bash
+# Find the config models
+grep -r "class.*BaseModel" src/ --include="*.py" -A 15
+```
+
+Check:
+- All config keys documented
+- Types and defaults match code
+- Config file search order is accurate
+- Example YAML would actually work
+
+### 4. Verify docs/architecture.md and CLAUDE.md
+
+```bash
+# What source files actually exist?
+git ls-files "src/**/*.py"
+```
+
+Check **both** `docs/architecture.md` and `CLAUDE.md` (Architecture section):
+- Listed files exist
+- No files are missing from the list
+- Descriptions match what the code does
+
+Both files have architecture listings that can drift independently.
+
+### 5. Check Examples
+
+For examples in any doc:
+- Would the YAML/commands actually work?
+- Are service names, paths, and options realistic?
+- Do examples use current syntax (not deprecated options)?
+
+### 6. Cross-Reference Consistency
+
+The same info appears in multiple places. Check for conflicts:
+- README.md vs docs/index.md
+- docs/commands.md vs CLAUDE.md command tables
+- Config examples across different docs
+
+### 7. Self-Check This Prompt
+
+This prompt can become outdated too. If you notice:
+- New automated checks that should be listed above
+- New doc files that need review guidelines
+- Patterns that caused issues
+
+Include prompt updates in your fixes.
+
+## Output Format
+
+Categorize findings:
+
+1. **Critical**: Wrong info that would break user workflows
+2. **Inaccuracy**: Technical errors (wrong defaults, paths, types)
+3. **Missing**: Undocumented features or options
+4. **Outdated**: Was true, no longer is
+5. **Inconsistency**: Docs contradict each other
+6. **Minor**: Typos, unclear wording
+
+For each issue, provide a ready-to-apply fix:
+
+```
+### Issue: [Brief description]
+
+- **File**: docs/commands.md:652
+- **Problem**: `cf ssh setup` has `--config` option but it's not documented
+- **Fix**: Add `--config, -c PATH` to the options table
+- **Verify**: `cf ssh setup --help`
+```

.prompts/duplication-audit.md

@@ -0,0 +1,79 @@
+# Duplication audit and generalization prompt
+
+You are a coding agent working inside a repository. Your job is to find duplicated
+functionality (not just identical code) and propose a minimal, safe generalization.
+Keep it simple and avoid adding features.
+
+## First steps
+
+- Read project-specific instructions (AGENTS.md, CONTRIBUTING.md, or similar) and follow them.
+- If instructions mention tooling or style (e.g., preferred search tools), use those.
+- Ask a brief clarification if the request is ambiguous (for example: report only vs refactor).
+
+## Objective
+
+Identify and consolidate duplicated functionality across the codebase. Duplication includes:
+- Multiple functions that parse or validate the same data in slightly different ways
+- Repeated file reads or config parsing
+- Similar command building or subprocess execution paths
+- Near-identical error handling or logging patterns
+- Repeated data transforms that can become a shared helper
+
+The goal is to propose a general, reusable abstraction that reduces duplication while
+preserving behavior. Keep changes minimal and easy to review.
+
+## Search strategy
+
+1) Map the hot paths
+- Scan entry points (CLI, web handlers, tasks, jobs) to see what they do repeatedly.
+- Look for cross-module patterns: same steps, different files.
+
+2) Find duplicate operations
+- Use fast search tools (prefer `rg`) to find repeated keywords and patterns.
+- Check for repeated YAML/JSON parsing, env interpolation, file IO, command building,
+  data validation, or response formatting.
+
+3) Validate duplication is real
+- Confirm the functional intent matches (not just similar code).
+- Note any subtle differences that must be preserved.
+
+4) Propose a minimal generalization
+- Suggest a shared helper, utility, or wrapper.
+- Avoid over-engineering. If only two call sites exist, keep the helper small.
+- Prefer pure functions and centralized IO if that already exists.
+
+## Deliverables
+
+Provide a concise report with:
+
+1) Findings
+- List duplicated behaviors with file references and a short description of the
+  shared functionality.
+- Explain why these are functionally the same (or nearly the same).
+
+2) Proposed generalizations
+- For each duplication, propose a shared helper and where it should live.
+- Outline any behavior differences that need to be parameterized.
+
+3) Impact and risk
+- Note any behavior risks, test needs, or migration steps.
+
+If the user asked you to implement changes:
+- Make only the minimal edits needed to dedupe behavior.
+- Keep the public API stable unless explicitly requested.
+- Add small comments only when the logic is non-obvious.
+- Summarize what changed and why.
+
+## Output format
+
+- Start with a short summary of the top 1-3 duplications.
+- Then provide a list of findings, ordered by impact.
+- Include a small proposed refactor plan (step-by-step, no more than 5 steps).
+- End with any questions or assumptions.
+
+## Guardrails
+
+- Do not add new features or change behavior beyond deduplication.
+- Avoid deep refactors without explicit request.
+- Preserve existing style conventions and import rules.
+- If a duplication is better left alone (e.g., clarity, single usage), say so.

.prompts/pr-review.md

@@ -0,0 +1,16 @@
+Review the pull request for:
+
+- **Code cleanliness**: Is the implementation clean and well-structured?
+- **DRY principle**: Does it avoid duplication?
+- **Code reuse**: Are there parts that should be reused from other places?
+- **Organization**: Is everything in the right place?
+- **Consistency**: Is it in the same style as other parts of the codebase?
+- **Simplicity**: Is it not over-engineered? Remember KISS and YAGNI. No dead code paths and NO defensive programming.
+- **No pointless wrappers**: Identify functions/methods that just call another function and return its result. Callers should call the underlying function directly instead of going through unnecessary indirection.
+- **User experience**: Does it provide a good user experience?
+- **PR**: Is the PR description and title clear and informative?
+- **Tests**: Are there tests, and do they cover the changes adequately? Are they testing something meaningful or are they just trivial?
+- **Live tests**: Test the changes in a REAL live environment to ensure they work as expected, use the config in `/opt/stacks/compose-farm.yaml`.
+- **Rules**: Does the code follow the project's coding standards and guidelines as laid out in @CLAUDE.md?
+
+Look at `git diff origin/main..HEAD` for the changes made in this pull request.

.prompts/update-demos.md

@@ -0,0 +1,51 @@
+Update demo recordings to match the current compose-farm.yaml configuration.
+
+## Key Gotchas
+
+1. **Never `git checkout` without asking** - check for uncommitted changes first
+2. **Prefer `nas` stacks** - demos run locally on nas, SSH adds latency
+3. **Terminal captures keyboard** - use `blur()` to release focus before command palette
+4. **Clicking sidebar navigates away** - clicking h1 scrolls to top
+5. **Buttons have icons, not text** - use `[data-tip="..."]` selectors
+6. **`record.py` auto-restores config** - no manual cleanup needed after CLI demos
+
+## Stacks Used in Demos
+
+| Stack | CLI Demos | Web Demos | Notes |
+|-------|-----------|-----------|-------|
+| `audiobookshelf` | quickstart, migration, apply | - | Migrates nas→anton |
+| `grocy` | update | navigation, stack, workflow, console | - |
+| `immich` | logs, compose | shell | Multiple containers |
+| `dozzle` | - | workflow | - |
+
+## CLI Demos
+
+**Files:** `docs/demos/cli/*.tape`
+
+Check:
+- `quickstart.tape`: `bat -r` line ranges match current config structure
+- `migration.tape`: nvim keystrokes work, stack exists on nas
+- `compose.tape`: exec commands produce meaningful output
+
+Run: `python docs/demos/cli/record.py [demo]`
+
+## Web Demos
+
+**Files:** `docs/demos/web/demo_*.py`
+
+Check:
+- Stack names in demos still exist in config
+- Selectors match current templates (grep for IDs in `templates/`)
+- Shell demo uses command palette for ALL navigation
+
+Run: `python docs/demos/web/record.py [demo]`
+
+## Before Recording
+
+```bash
+# Check for uncommitted config changes
+git -C /opt/stacks diff compose-farm.yaml
+
+# Verify stacks are running
+cf ps audiobookshelf grocy immich dozzle
+```

CLAUDE.md

@@ -9,20 +9,88 @@
 ## Architecture
 
 ```
-compose_farm/
-├── config.py  # Pydantic models, YAML loading
-├── ssh.py     # asyncssh execution, streaming
-└── cli.py     # Typer commands
+src/compose_farm/
+├── cli/               # CLI subpackage
+│   ├── __init__.py    # Imports modules to trigger command registration
+│   ├── app.py         # Shared Typer app instance, version callback
+│   ├── common.py      # Shared helpers, options, progress bar utilities
+│   ├── config.py      # Config subcommand (init, show, path, validate, edit, symlink)
+│   ├── lifecycle.py   # up, down, stop, pull, restart, update, apply, compose commands
+│   ├── management.py  # refresh, check, init-network, traefik-file commands
+│   ├── monitoring.py  # logs, ps, stats, list commands
+│   ├── ssh.py         # SSH key management (setup, status, keygen)
+│   └── web.py         # Web UI server command
+├── compose.py         # Compose file parsing (.env, ports, volumes, networks)
+├── config.py          # Pydantic models, YAML loading
+├── console.py         # Shared Rich console instances
+├── executor.py        # SSH/local command execution, streaming output
+├── glances.py         # Glances API integration for host resource stats
+├── logs.py            # Image digest snapshots (dockerfarm-log.toml)
+├── operations.py      # Business logic (up, migrate, discover, preflight checks)
+├── paths.py           # Path utilities, config file discovery
+├── registry.py        # Container registry client for update checking
+├── ssh_keys.py        # SSH key path constants and utilities
+├── state.py           # Deployment state tracking (which stack on which host)
+├── traefik.py         # Traefik file-provider config generation from labels
+└── web/               # Web UI (FastAPI + HTMX)
 ```
 
+## Web UI Icons
+
+Icons use [Lucide](https://lucide.dev/). Add new icons as macros in `web/templates/partials/icons.html` by copying SVG paths from their site. The `action_btn`, `stat_card`, and `collapse` macros in `components.html` accept an optional `icon` parameter.
+
+## HTMX Patterns
+
+- **Multi-element refresh**: Use custom events, not `hx-swap-oob`. Elements have `hx-trigger="cf:refresh from:body"` and JS calls `document.body.dispatchEvent(new CustomEvent('cf:refresh'))`. Simpler to debug/test.
+- **SPA navigation**: Sidebar uses `hx-boost="true"` to AJAX-ify links.
+- **Attribute inheritance**: Set `hx-target`/`hx-swap` on parent elements.
+
 ## Key Design Decisions
 
-1. **asyncssh over Paramiko/Fabric**: Native async support, built-in streaming
-2. **Parallel by default**: Multiple services run concurrently via `asyncio.gather`
-3. **Streaming output**: Real-time stdout/stderr with `[service]` prefix
+1. **Hybrid SSH approach**: asyncssh for parallel streaming with prefixes; native `ssh -t` for raw mode (progress bars)
+2. **Parallel by default**: Multiple stacks run concurrently via `asyncio.gather`
+3. **Streaming output**: Real-time stdout/stderr with `[stack]` prefix using Rich
 4. **SSH key auth only**: Uses ssh-agent, no password handling (YAGNI)
 5. **NFS assumption**: Compose files at same path on all hosts
-6. **Local execution**: When host is `localhost`/`local`, skip SSH and run locally
+6. **Local IP auto-detection**: Skips SSH when target host matches local machine's IP
+7. **State tracking**: Tracks where stacks are deployed for auto-migration
+8. **Pre-flight checks**: Verifies NFS mounts and Docker networks exist before starting/migrating
+
+## Code Style
+
+- **Imports at top level**: Never add imports inside functions unless they are explicitly marked with `# noqa: PLC0415` and a comment explaining it speeds up CLI startup. Heavy modules like `pydantic`, `yaml`, and `rich.table` are lazily imported to keep `cf --help` fast.
+
+## Development Commands
+
+Use `just` for common tasks. Run `just` to list available commands:
+
+| Command | Description |
+|---------|-------------|
+| `just install` | Install dev dependencies |
+| `just test` | Run all tests |
+| `just test-cli` | Run CLI tests (parallel) |
+| `just test-web` | Run web UI tests (parallel) |
+| `just lint` | Lint, format, and type check |
+| `just web` | Start web UI (port 9001) |
+| `just doc` | Build and serve docs (port 9002) |
+| `just clean` | Clean build artifacts |
+
+## Testing
+
+Run tests with `just test` or `uv run pytest`. Browser tests require Chromium (system-installed or via `playwright install chromium`):
+
+```bash
+# Unit tests only (parallel)
+uv run pytest -m "not browser" -n auto
+
+# Browser tests only (parallel)
+uv run pytest -m browser -n auto
+
+# All tests
+uv run pytest
+```
+
+Browser tests are marked with `@pytest.mark.browser`. They use Playwright to test HTMX behavior, JavaScript functionality (sidebar filter, command palette, terminals), and content stability during navigation.
 
 ## Communication Notes
 
@@ -34,14 +102,66 @@ compose_farm/
 - **NEVER merge anything into main.** Always commit directly or use fast-forward/rebase.
 - Never force push.
 
+## SSH Agent in Remote Sessions
+
+When pushing to GitHub via SSH fails with "Permission denied (publickey)", fix the SSH agent socket:
+
+```bash
+# Find and set the correct SSH agent socket
+SSH_AUTH_SOCK=$(ls -t ~/.ssh/agent/s.*.sshd.* 2>/dev/null | head -1) git push origin branch-name
+```
+
+This is needed because the SSH agent socket path changes between sessions.
+
+## Pull Requests
+
+- Never include unchecked checklists (e.g., `- [ ] ...`) in PR descriptions. Either omit the checklist or use checked items.
+- **NEVER run `gh pr merge`**. PRs are merged via the GitHub UI, not the CLI.
+
+## Releases
+
+Use `gh release create` to create releases. The tag is created automatically.
+
+```bash
+# IMPORTANT: Ensure you're on latest origin/main before releasing!
+git fetch origin
+git checkout origin/main
+
+# Check current version
+git tag --sort=-v:refname | head -1
+
+# Create release (minor version bump: v0.21.1 -> v0.22.0)
+gh release create v0.22.0 --title "v0.22.0" --notes "release notes here"
+```
+
+Versioning:
+- **Patch** (v0.21.0 → v0.21.1): Bug fixes
+- **Minor** (v0.21.1 → v0.22.0): New features, non-breaking changes
+
+Write release notes manually describing what changed. Group by features and bug fixes.
+
 ## Commands Quick Reference
 
-| Command | Docker Compose Equivalent |
-|---------|--------------------------|
-| `up`    | `docker compose up -d`   |
-| `down`  | `docker compose down`    |
-| `pull`  | `docker compose pull`    |
-| `restart` | `down` + `up -d`       |
-| `update` | `pull` + `down` + `up -d` |
-| `logs`  | `docker compose logs`    |
-| `ps`    | `docker compose ps`      |
+CLI available as `cf` or `compose-farm`.
+
+| Command | Description |
+|---------|-------------|
+| `up`    | Start stacks (`docker compose up -d`), auto-migrates if host changed |
+| `down`  | Stop stacks (`docker compose down`). Use `--orphaned` to stop stacks removed from config |
+| `stop`  | Stop services without removing containers (`docker compose stop`) |
+| `pull`  | Pull latest images |
+| `restart` | Restart running containers (`docker compose restart`) |
+| `update` | Pull, build, recreate only if changed (`up -d --pull always --build`) |
+| `apply` | Make reality match config: migrate stacks + stop orphans. Use `--dry-run` to preview |
+| `compose` | Run any docker compose command on a stack (passthrough) |
+| `logs`  | Show stack logs |
+| `ps`    | Show status of all stacks |
+| `stats` | Show overview (hosts, stacks, pending migrations; `--live` for container counts) |
+| `list`  | List stacks and hosts (`--simple` for scripting, `--host` to filter) |
+| `refresh` | Update state from reality: discover running stacks, capture image digests |
+| `check` | Validate config, traefik labels, mounts, networks; show host compatibility |
+| `init-network` | Create Docker network on hosts with consistent subnet/gateway |
+| `traefik-file` | Generate Traefik file-provider config from compose labels |
+| `config` | Manage config files (init, init-env, show, path, validate, edit, symlink) |
+| `ssh`   | Manage SSH keys (setup, status, keygen) |
+| `web`   | Start web UI server |

Dockerfile

@@ -0,0 +1,28 @@
+# syntax=docker/dockerfile:1
+
+# Build stage - install with uv
+FROM ghcr.io/astral-sh/uv:python3.14-alpine AS builder
+
+ARG VERSION
+RUN uv tool install --compile-bytecode "compose-farm[web]${VERSION:+==$VERSION}"
+
+# Runtime stage - minimal image without uv
+FROM python:3.14-alpine
+
+# Install only runtime requirements
+RUN apk add --no-cache openssh-client
+
+# Copy installed tool virtualenv and bin symlinks from builder
+COPY --from=builder /root/.local/share/uv/tools/compose-farm /root/.local/share/uv/tools/compose-farm
+COPY --from=builder /usr/local/bin/cf /usr/local/bin/compose-farm /usr/local/bin/
+
+# Allow non-root users to access the installed tool
+# (required when running with user: "${CF_UID:-0}:${CF_GID:-0}")
+RUN chmod 755 /root
+
+# Allow non-root users to add passwd entries (required for SSH)
+RUN chmod 666 /etc/passwd
+
+# Entrypoint creates /etc/passwd entry for non-root UIDs (required for SSH)
+ENTRYPOINT ["sh", "-c", "[ $(id -u) != 0 ] && echo ${USER:-u}:x:$(id -u):$(id -g)::${HOME:-/}:/bin/sh >> /etc/passwd; exec cf \"$@\"", "--"]
+CMD ["--help"]

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 Bas Nijholt
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

PLAN.md

@@ -1,35 +0,0 @@
-# Compose Farm – Traefik Multihost Ingress Plan
-
-## Goal
-Generate a Traefik file-provider fragment from existing docker-compose Traefik labels (no config duplication) so a single front-door Traefik on 192.168.1.66 with wildcard `*.lab.mydomain.org` can route to services running on other hosts. Keep the current simplicity (SSH + docker compose); no Swarm/K8s.
-
-## Requirements
-- Traefik stays on main host; keep current `dynamic.yml` and Docker provider for local containers.
-- Add a watched directory provider (any path works) and load a generated fragment (e.g., `compose-farm.generated.yml`).
-- No edits to compose files: reuse existing `traefik.*` labels as the single source of truth; Compose Farm only reads them.
-- Generator infers routing from labels and reachability from `ports:` mappings; prefer host-published ports so Traefik can reach services across hosts. Upstreams point to `<host address>:<published host port>`; warn if no published port is found.
-- Only minimal data in `compose-farm.yaml`: hosts map and service→host mapping (already present).
-- No new orchestration/discovery layers; respect KISS/YAGNI/DRY.
-
-## Non-Goals
-- No Swarm/Kubernetes adoption.
-- No global Docker provider across hosts.
-- No health checks/service discovery layer.
-
-## Current State (Dec 2025)
-- Compose Farm: Typer CLI wrapping `docker compose` over SSH; config in `compose-farm.yaml`; parallel by default; snapshot/log tooling present.
-- Traefik: single instance on 192.168.1.66, wildcard `*.lab.mydomain.org`, Docker provider for local services, file provider via `dynamic.yml` already in use.
-
-## Proposed Implementation Steps
-1) Add generator command: `compose-farm traefik-file --output <path>`.
-2) Resolve per-service host from `compose-farm.yaml`; read compose file at `{compose_dir}/{service}/docker-compose.yml`.
-3) Parse `traefik.*` labels to build routers/services/middlewares as in compose; map container port to published host port (from `ports:`) to form upstream URLs with host address.
-4) Emit file-provider YAML to the watched directory (recommended default: `/mnt/data/traefik/dynamic.d/compose-farm.generated.yml`, but user chooses via `--output`).
-5) Warnings: if no published port is found, warn that cross-host reachability requires L3 reachability to container IPs.
-6) Tests: label parsing, port mapping, YAML render; scenario with published port; scenario without published port.
-7) Docs: update README/CLAUDE to describe directory provider flags and the generator workflow; note that compose files remain unchanged.
-
-## Open Questions
-- How to derive target host address: use `hosts.<name>.address` verbatim, or allow override per service? (Default: use host address.)
-- Should we support multiple hosts/backends per service for LB/HA? (Start with single server.)
-- Where to store generated file by default? (Default to user-specified `--output`; maybe fallback to `./compose-farm-traefik.yml`.)

compose-farm.example.yaml

@@ -3,23 +3,28 @@
 
 compose_dir: /opt/compose
 
+# Optional: Auto-regenerate Traefik file-provider config after up/down/update
+traefik_file: /opt/traefik/dynamic.d/compose-farm.yml
+traefik_stack: traefik  # Skip stacks on same host (docker provider handles them)
+
 hosts:
   # Full form with all options
-  nas01:
+  server-1:
     address: 192.168.1.10
     user: docker
     port: 22
 
   # Short form (just address, user defaults to current user)
-  nas02: 192.168.1.11
+  server-2: 192.168.1.11
 
   # Local execution (no SSH)
   local: localhost
 
-services:
-  # Map service names to hosts
-  # Compose file expected at: {compose_dir}/{service}/docker-compose.yml
-  plex: nas01
-  jellyfin: nas02
-  sonarr: nas01
-  radarr: nas02
+stacks:
+  # Map stack names to hosts
+  # Compose file expected at: {compose_dir}/{stack}/compose.yaml
+  traefik: server-1    # Traefik runs here
+  plex: server-2       # Stacks on other hosts get file-provider entries
+  jellyfin: server-2
+  grafana: server-1
+  nextcloud: local

docker-compose.yml

@@ -0,0 +1,70 @@
+services:
+  cf:
+    image: ghcr.io/basnijholt/compose-farm:latest
+    # Run as current user to preserve file ownership on mounted volumes
+    # Set CF_UID=$(id -u) CF_GID=$(id -g) in your environment or .env file
+    # Defaults to root (0:0) for backwards compatibility
+    user: "${CF_UID:-0}:${CF_GID:-0}"
+    volumes:
+      # Compose directory (contains compose files AND compose-farm.yaml config)
+      - ${CF_COMPOSE_DIR:-/opt/stacks}:${CF_COMPOSE_DIR:-/opt/stacks}
+      # SSH keys for passwordless auth (generated by `cf ssh setup`)
+      # Choose ONE option below (use the same option for both cf and web services):
+      # Option 1: Host path (default) - keys at ~/.ssh/compose-farm/id_ed25519
+      - ${CF_SSH_DIR:-~/.ssh/compose-farm}:${CF_HOME:-/root}/.ssh/compose-farm
+      # Option 2: Named volume - managed by Docker, shared between services
+      # - cf-ssh:${CF_HOME:-/root}/.ssh
+      # Option 3: SSH agent forwarding (uncomment if using ssh-agent)
+      # - ${SSH_AUTH_SOCK}:/ssh-agent:ro
+    environment:
+      - SSH_AUTH_SOCK=/ssh-agent
+      # Config file path (state stored alongside it)
+      - CF_CONFIG=${CF_COMPOSE_DIR:-/opt/stacks}/compose-farm.yaml
+      # HOME must match the user running the container for SSH to find keys
+      - HOME=${CF_HOME:-/root}
+      # USER is required for SSH when running as non-root (UID not in /etc/passwd)
+      - USER=${CF_USER:-root}
+
+  web:
+    image: ghcr.io/basnijholt/compose-farm:latest
+    restart: unless-stopped
+    command: web --host 0.0.0.0 --port 9000
+    # Run as current user to preserve file ownership on mounted volumes
+    user: "${CF_UID:-0}:${CF_GID:-0}"
+    volumes:
+      - ${CF_COMPOSE_DIR:-/opt/stacks}:${CF_COMPOSE_DIR:-/opt/stacks}
+      # SSH keys - use the SAME option as cf service above
+      # Option 1: Host path (default)
+      - ${CF_SSH_DIR:-~/.ssh/compose-farm}:${CF_HOME:-/root}/.ssh/compose-farm
+      # Option 2: Named volume
+      # - cf-ssh:${CF_HOME:-/root}/.ssh
+      # Option 3: SSH agent forwarding (uncomment if using ssh-agent)
+      # - ${SSH_AUTH_SOCK}:/ssh-agent:ro
+      # XDG config dir for backups and image digest logs (persists across restarts)
+      - ${CF_XDG_CONFIG:-~/.config/compose-farm}:${CF_HOME:-/root}/.config/compose-farm
+    environment:
+      - SSH_AUTH_SOCK=/ssh-agent
+      - CF_CONFIG=${CF_COMPOSE_DIR:-/opt/stacks}/compose-farm.yaml
+      # Used to detect self-updates and run via SSH to survive container restart
+      - CF_WEB_STACK=compose-farm
+      # HOME must match the user running the container for SSH to find keys
+      - HOME=${CF_HOME:-/root}
+      # USER is required for SSH when running as non-root (UID not in /etc/passwd)
+      - USER=${CF_USER:-root}
+    labels:
+      - traefik.enable=true
+      - traefik.http.routers.compose-farm.rule=Host(`compose-farm.${DOMAIN}`)
+      - traefik.http.routers.compose-farm.entrypoints=websecure
+      - traefik.http.routers.compose-farm-local.rule=Host(`compose-farm.local`)
+      - traefik.http.routers.compose-farm-local.entrypoints=web
+      - traefik.http.services.compose-farm.loadbalancer.server.port=9000
+    networks:
+      - mynetwork
+
+networks:
+  mynetwork:
+    external: true
+
+volumes:
+  cf-ssh:
+    # Only used if Option 2 is selected above

docs/CNAME

@@ -0,0 +1 @@
+compose-farm.nijho.lt

docs/architecture.md

@@ -0,0 +1,361 @@
+---
+icon: lucide/layers
+---
+
+# Architecture
+
+This document explains how Compose Farm works under the hood.
+
+## Design Philosophy
+
+Compose Farm follows three core principles:
+
+1. **KISS** - Keep it simple. It's a thin wrapper around `docker compose` over SSH.
+2. **YAGNI** - No orchestration, no service discovery, no health checks until needed.
+3. **Zero changes** - Your existing compose files work unchanged.
+
+## High-Level Architecture
+
+```
+┌─────────────────────────────────────────────────────────────────┐
+│                        Compose Farm CLI                         │
+│                                                                 │
+│  ┌──────────┐  ┌──────────┐  ┌──────────┐  ┌──────────────────┐ │
+│  │  Config  │  │  State   │  │Operations│  │   Executor       │ │
+│  │  Parser  │  │ Tracker  │  │  Logic   │  │  (SSH/Local)     │ │
+│  └────┬─────┘  └────┬─────┘  └────┬─────┘  └────────┬─────────┘ │
+└───────┼─────────────┼─────────────┼─────────────────┼───────────┘
+        │             │             │                 │
+        ▼             ▼             ▼                 ▼
+┌───────────────────────────────────────────────────────────────┐
+│                         SSH / Local                            │
+└───────────────────────────────────────────────────────────────┘
+        │                                             │
+        ▼                                             ▼
+┌───────────────┐                           ┌───────────────┐
+│   Host: nuc   │                           │   Host: hp    │
+│               │                           │               │
+│ docker compose│                           │ docker compose│
+│    up -d      │                           │    up -d      │
+└───────────────┘                           └───────────────┘
+```
+
+## Core Components
+
+### Configuration (`src/compose_farm/config.py`)
+
+Pydantic models for YAML configuration:
+
+- **Config** - Root configuration with compose_dir, hosts, stacks
+- **Host** - Host address, SSH user, and port
+
+Key features:
+- Validation with Pydantic
+- Multi-host stack expansion (`all` → list of hosts)
+- YAML loading with sensible defaults
+
+### State Tracking (`src/compose_farm/state.py`)
+
+Tracks deployment state in `compose-farm-state.yaml` (stored alongside the config file):
+
+```yaml
+deployed:
+  plex: nuc
+  grafana: nuc
+```
+
+Used for:
+- Detecting migrations (stack moved to different host)
+- Identifying orphans (stacks removed from config)
+- `cf ps` status display
+
+### Operations (`src/compose_farm/operations.py`)
+
+Business logic for stack operations:
+
+- **up** - Start stack, handle migration if needed
+- **down** - Stop stack
+- **preflight checks** - Verify mounts, networks exist before operations
+- **discover** - Find running stacks on hosts
+- **migrate** - Down on old host, up on new host
+
+### Executor (`src/compose_farm/executor.py`)
+
+SSH and local command execution:
+
+- **Hybrid SSH approach**: asyncssh for parallel streaming, native `ssh -t` for raw mode
+- **Parallel by default**: Multiple stacks via `asyncio.gather`
+- **Streaming output**: Real-time stdout/stderr with `[stack]` prefix
+- **Local detection**: Skips SSH when target matches local machine IP
+
+### CLI (`src/compose_farm/cli/`)
+
+Typer-based CLI with subcommand modules:
+
+```
+cli/
+├── app.py          # Shared Typer app, version callback
+├── common.py       # Shared helpers, options, progress utilities
+├── config.py       # config subcommand (init, init-env, show, path, validate, edit, symlink)
+├── lifecycle.py    # up, down, stop, pull, restart, update, apply, compose
+├── management.py   # refresh, check, init-network, traefik-file
+├── monitoring.py   # logs, ps, stats
+├── ssh.py          # SSH key management (setup, status, keygen)
+└── web.py          # Web UI server command
+```
+
+## Command Flow
+
+### cf up plex
+
+```
+1. Load configuration
+   └─► Parse compose-farm.yaml
+   └─► Validate stack exists
+
+2. Check state
+   └─► Load state.yaml
+   └─► Is plex already running?
+   └─► Is it on a different host? (migration needed)
+
+3. Pre-flight checks
+   └─► SSH to target host
+   └─► Check compose file exists
+   └─► Check required mounts exist
+   └─► Check required networks exist
+
+4. Execute migration (if needed)
+   └─► SSH to old host
+   └─► Run: docker compose down
+
+5. Start stack
+   └─► SSH to target host
+   └─► cd /opt/compose/plex
+   └─► Run: docker compose up -d
+
+6. Update state
+   └─► Write new state to state.yaml
+
+7. Generate Traefik config (if configured)
+   └─► Regenerate traefik file-provider
+```
+
+### cf apply
+
+```
+1. Load configuration and state
+
+2. Compute diff
+   ├─► Orphans: in state, not in config
+   ├─► Migrations: in both, different host
+   └─► Missing: in config, not in state
+
+3. Stop orphans
+   └─► For each orphan: cf down
+
+4. Migrate stacks
+   └─► For each migration: down old, up new
+
+5. Start missing
+   └─► For each missing: cf up
+
+6. Update state
+```
+
+## SSH Execution
+
+### Parallel Streaming (asyncssh)
+
+For most operations, Compose Farm uses asyncssh:
+
+```python
+async def run_command(host, command):
+    async with asyncssh.connect(host) as conn:
+        result = await conn.run(command)
+        return result.stdout, result.stderr
+```
+
+Multiple stacks run concurrently via `asyncio.gather`.
+
+### Raw Mode (native ssh)
+
+For commands needing PTY (progress bars, interactive):
+
+```bash
+ssh -t user@host "docker compose pull"
+```
+
+### Local Detection
+
+When target host IP matches local machine:
+
+```python
+if is_local(host_address):
+    # Run locally, no SSH
+    subprocess.run(command)
+else:
+    # SSH to remote
+    ssh.run(command)
+```
+
+## State Management
+
+### State File
+
+Location: `compose-farm-state.yaml` (stored alongside the config file)
+
+```yaml
+deployed:
+  plex: nuc
+  grafana: nuc
+```
+
+Image digests are stored separately in `dockerfarm-log.toml` (also in the config directory).
+
+### State Transitions
+
+```
+Config Change          State Change           Action
+─────────────────────────────────────────────────────
+Add stack            Missing                 cf up
+Remove stack         Orphaned                cf down
+Change host           Migration               down old, up new
+No change             No change               none (or refresh)
+```
+
+### cf refresh
+
+Syncs state with reality by querying Docker on each host:
+
+```bash
+docker ps --format '{{.Names}}'
+```
+
+Updates state.yaml to match what's actually running.
+
+## Compose File Discovery
+
+For each stack, Compose Farm looks for compose files in:
+
+```
+{compose_dir}/{stack}/
+├── compose.yaml         # preferred
+├── compose.yml
+├── docker-compose.yml
+└── docker-compose.yaml
+```
+
+First match wins.
+
+## Traefik Integration
+
+### Label Extraction
+
+Compose Farm parses Traefik labels from compose files:
+
+```yaml
+stacks:
+  plex:
+    labels:
+      - traefik.enable=true
+      - traefik.http.routers.plex.rule=Host(`plex.example.com`)
+      - traefik.http.services.plex.loadbalancer.server.port=32400
+```
+
+### File Provider Generation
+
+Converts labels to Traefik file-provider YAML:
+
+```yaml
+http:
+  routers:
+    plex:
+      rule: Host(`plex.example.com`)
+      service: plex
+  services:
+    plex:
+      loadBalancer:
+        servers:
+          - url: http://192.168.1.10:32400
+```
+
+### Variable Resolution
+
+Supports `${VAR}` and `${VAR:-default}` from:
+1. Service's `.env` file
+2. Current environment
+
+## Error Handling
+
+### Pre-flight Failures
+
+Before any operation, Compose Farm checks:
+- SSH connectivity
+- Compose file existence
+- Required mounts
+- Required networks
+
+If checks fail, operation aborts with clear error.
+
+### Partial Failures
+
+When operating on multiple stacks:
+- Each stack is independent
+- Failures are logged, but other stacks continue
+- Exit code reflects overall success/failure
+
+## Performance Considerations
+
+### Parallel Execution
+
+Services are started/stopped in parallel:
+
+```python
+await asyncio.gather(*[
+    up_stack(stack) for stack in stacks
+])
+```
+
+### SSH Multiplexing
+
+For repeated connections to the same host, SSH reuses connections.
+
+### Caching
+
+- Config is parsed once per command
+- State is loaded once, written once
+- Host discovery results are cached during command
+
+## Web UI Architecture
+
+```
+┌─────────────────────────────────────────────────────────────┐
+│                         Web UI                               │
+│                                                             │
+│  ┌─────────────┐  ┌─────────────┐  ┌─────────────────────┐  │
+│  │   FastAPI   │  │    Jinja    │  │       HTMX          │  │
+│  │   Backend   │  │  Templates  │  │   Dynamic Updates   │  │
+│  └─────────────┘  └─────────────┘  └─────────────────────┘  │
+│                                                             │
+│  Pattern: Custom events, not hx-swap-oob                    │
+│  Elements trigger on: cf:refresh from:body                  │
+└─────────────────────────────────────────────────────────────┘
+```
+
+Icons use [Lucide](https://lucide.dev/). Add new icons as macros in `web/templates/partials/icons.html`.
+
+### Host Resource Monitoring (`src/compose_farm/glances.py`)
+
+Integration with [Glances](https://nicolargo.github.io/glances/) for real-time host stats:
+
+- Fetches CPU, memory, and load from Glances REST API on each host
+- Used by web UI dashboard to display host resource usage
+- Requires `glances_stack` config option pointing to a Glances stack running on all hosts
+
+### Container Registry Client (`src/compose_farm/registry.py`)
+
+OCI Distribution API client for checking image updates:
+
+- Parses image references (registry, namespace, name, tag, digest)
+- Fetches available tags from Docker Hub, GHCR, and other registries
+- Compares semantic versions to find newer releases

docs/assets/apply.gif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01dabdd8f62773823ba2b8dc74f9931f1a1b88215117e6a080004096025491b0
+size 901456

docs/assets/apply.webm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:134c903a6b3acfb933617b33755b0cdb9bac2a59e5e35b64236e248a141d396d
+size 206883

docs/assets/compose.gif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8b3cdb3486ec79b3ddb2f7571c13d54ac9aed182edfe708eff76a966a90cfc7
+size 1132310

docs/assets/compose.webm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3c4d4a62f062f717df4e6752efced3caea29004dc90fe97fd7633e7f0ded9db
+size 341057

docs/assets/install.gif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c1bb48cc2f364681515a4d8bd0c586d133f5a32789b7bb64524ad7d9ed0a8e9
+size 543135

docs/assets/install.webm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f82d96137f039f21964c15c1550aa1b1f0bb2d52c04d012d253dbfbd6fad096
+size 268086

docs/assets/logs.gif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a4045b00d90928f42c7764b3c24751576cfb68a34c6e84d12b4e282d2e67378
+size 146467

docs/assets/logs.webm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1b94416ed3740853f863e19bf45f26241a203fb0d7d187160a537f79aa544fa
+size 60353

docs/assets/migration.gif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:848d9c48fb7511da7996149277c038589fad1ee406ff2f30c28f777fc441d919
+size 1183641

docs/assets/migration.webm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e747ee71bb38b19946005d5a4def4d423dadeaaade452dec875c4cb2d24a5b77
+size 407373

docs/assets/quickstart.gif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d32c9a3eec06e57df085ad347e6bf61e323f8bd8322d0c540f0b9d4834196dfd
+size 3589776

docs/assets/quickstart.webm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c54eda599389dac74c24c83527f95cd1399e653d7faf2972c2693d90e590597
+size 1085344

docs/assets/update.gif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62f9b5ec71496197a3f1c3e3bca8967d603838804279ea7dbf00a70d3391ff6c
+size 127123

docs/assets/update.webm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac2b93d3630af87b44a135723c5d10e8287529bed17c28301b2802cd9593e9e8
+size 98748

docs/assets/web-console.gif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b50a7e9836c496c0989363d1440fa0a6ccdaa38ee16aae92b389b3cf3c3732f
+size 2385110

docs/assets/web-console.webm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccbb3d5366c7734377e12f98cca0b361028f5722124f1bb7efa231f6aeffc116
+size 2208044

docs/assets/web-live_stats.gif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4135888689a10c5ae2904825d98f2a6d215c174a4bd823e25761f619590f04ff
+size 3990104

docs/assets/web-live_stats.webm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87739cd6f6576a81100392d8d1e59d3e776fecc8f0721a31332df89e7fc8593d
+size 5814274

docs/assets/web-navigation.gif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:269993b52721ce70674d3aab2a4cd8c58aa621d4ba0739afedae661c90965b26
+size 3678371

docs/assets/web-navigation.webm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0098b55bb6a52fa39f807a01fa352ce112bcb446e2a2acb963fb02d21b28c934
+size 3088813

docs/assets/web-shell.gif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4bf9d8c247d278799d1daea784fc662a22f12b1bd7883f808ef30f35025ebca6
+size 4166443

docs/assets/web-shell.webm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02d5124217a94849bf2971d6d13d28da18c557195a81b9cca121fb7c07f0501b
+size 3523244

docs/assets/web-stack.gif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:412a0e68f8e52801cafbb9a703ca9577e7c14cc7c0e439160b9185961997f23c
+size 4435697

docs/assets/web-stack.webm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e600a1d3216b44497a889f91eac94d62ef7207b4ed0471465dcb72408caa28e
+size 3764693

docs/assets/web-themes.gif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c07a283f4f70c4ab205b0f0acb5d6f55e3ced4c12caa7a8d5914ffe3548233a
+size 5768166

docs/assets/web-themes.webm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:562228841de976d70ee80999b930eadf3866a13ff2867d900279993744c44671
+size 6667918

docs/assets/web-workflow.gif

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:845746ac1cb101c3077d420c4f3fda3ca372492582dc123ac8a031a68ae9b6b1
+size 12943150

docs/assets/web-workflow.webm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:189259558b5760c02583885168d7b0b47cf476cba81c7c028ec770f9d6033129
+size 12415357

docs/best-practices.md

@@ -0,0 +1,372 @@
+---
+icon: lucide/lightbulb
+---
+
+# Best Practices
+
+Tips, limitations, and recommendations for using Compose Farm effectively.
+
+## Limitations
+
+### No Cross-Host Networking
+
+Compose Farm moves containers between hosts but **does not provide cross-host networking**. Docker's internal DNS and networks don't span hosts.
+
+**What breaks when you move a stack:**
+
+| Feature | Works? | Why |
+|---------|--------|-----|
+| `http://redis:6379` | No | Docker DNS doesn't cross hosts |
+| Docker network names | No | Networks are per-host |
+| `DATABASE_URL=postgres://db:5432` | No | Container name won't resolve |
+| Host IP addresses | Yes | Use `192.168.1.10:5432` |
+
+### What Compose Farm Doesn't Do
+
+- No overlay networking (use Swarm/Kubernetes)
+- No service discovery across hosts
+- No automatic dependency tracking between compose files
+- No health checks or restart policies beyond Docker's
+- No secrets management beyond Docker's
+
+## Stack Organization
+
+### Keep Dependencies Together
+
+If services talk to each other, keep them in the same compose file on the same host:
+
+```yaml
+# /opt/compose/myapp/docker-compose.yml
+services:
+  app:
+    image: myapp
+    depends_on:
+      - db
+      - redis
+
+  db:
+    image: postgres
+
+  redis:
+    image: redis
+```
+
+```yaml
+# compose-farm.yaml
+stacks:
+  myapp: nuc  # All three containers stay together
+```
+
+### Separate Standalone Stacks
+
+Stacks whose services don't talk to other containers can be anywhere:
+
+```yaml
+stacks:
+  # These can run on any host
+  plex: nuc
+  jellyfin: hp
+  homeassistant: nas
+
+  # These should stay together
+  myapp: nuc  # includes app + db + redis
+```
+
+### Cross-Host Communication
+
+If services MUST communicate across hosts, publish ports:
+
+```yaml
+# Instead of
+DATABASE_URL=postgres://db:5432
+
+# Use
+DATABASE_URL=postgres://192.168.1.10:5432
+```
+
+```yaml
+# And publish the port
+services:
+  db:
+    ports:
+      - "5432:5432"
+```
+
+## Multi-Host Stacks
+
+### When to Use `all`
+
+Use `all` for stacks that need local access to each host:
+
+```yaml
+stacks:
+  # Need Docker socket
+  dozzle: all          # Log viewer
+  portainer-agent: all  # Portainer agents
+  autokuma: all        # Auto-creates monitors
+
+  # Need host metrics
+  node-exporter: all   # Prometheus metrics
+  promtail: all        # Log shipping
+```
+
+### Host-Specific Lists
+
+For stacks on specific hosts only:
+
+```yaml
+stacks:
+  # Only on compute nodes
+  gitlab-runner: [nuc, hp]
+
+  # Only on storage nodes
+  minio: [nas-1, nas-2]
+```
+
+## Migration Safety
+
+### Pre-flight Checks
+
+Before migrating, Compose Farm verifies:
+- Compose file is accessible on new host
+- Required mounts exist on new host
+- Required networks exist on new host
+
+### Data Considerations
+
+**Compose Farm doesn't move data.** Ensure:
+
+1. **Shared storage**: Data volumes on NFS/shared storage
+2. **External databases**: Data in external DB, not container
+3. **Backup first**: Always backup before migration
+
+### Safe Migration Pattern
+
+```bash
+# 1. Preview changes
+cf apply --dry-run
+
+# 2. Verify target host can run the stack
+cf check myservice
+
+# 3. Apply changes
+cf apply
+```
+
+## State Management
+
+### When to Refresh
+
+Run `cf refresh` after:
+- Manual `docker compose` commands
+- Container restarts
+- Host reboots
+- Any changes outside Compose Farm
+
+```bash
+cf refresh --dry-run  # Preview
+cf refresh            # Sync
+```
+
+### State Conflicts
+
+If state doesn't match reality:
+
+```bash
+# See what's actually running
+cf refresh --dry-run
+
+# Sync state
+cf refresh
+
+# Then apply config
+cf apply
+```
+
+## Shared Storage
+
+### NFS Best Practices
+
+```bash
+# Mount options for Docker compatibility
+nas:/compose /opt/compose nfs rw,hard,intr,rsize=8192,wsize=8192 0 0
+```
+
+### Directory Ownership
+
+Ensure consistent UID/GID across hosts:
+
+```yaml
+services:
+  myapp:
+    environment:
+      - PUID=1000
+      - PGID=1000
+```
+
+### Config vs Data
+
+Keep config and data separate:
+
+```
+/opt/compose/          # Shared: compose files + config
+├── plex/
+│   ├── docker-compose.yml
+│   └── config/        # Small config files OK
+
+/mnt/data/             # Shared: large media files
+├── movies/
+├── tv/
+└── music/
+
+/opt/appdata/          # Local: per-host app data
+├── plex/
+└── grafana/
+```
+
+## Performance
+
+### Parallel Operations
+
+Compose Farm runs operations in parallel. For large deployments:
+
+```bash
+# Good: parallel by default
+cf up --all
+
+# Avoid: sequential updates when possible
+for svc in plex grafana nextcloud; do
+  cf update $svc
+done
+```
+
+### SSH Connection Reuse
+
+SSH connections are reused within a command. For many operations:
+
+```bash
+# One command, one connection per host
+cf update --all
+
+# Multiple commands, multiple connections (slower)
+cf update plex && cf update grafana && cf update nextcloud
+```
+
+## Traefik Setup
+
+### Stack Placement
+
+Put Traefik on a reliable host:
+
+```yaml
+stacks:
+  traefik: nuc  # Primary host with good uptime
+```
+
+### Same-Host Stacks
+
+Stacks on the same host as Traefik use Docker provider:
+
+```yaml
+traefik_stack: traefik
+
+stacks:
+  traefik: nuc
+  portainer: nuc   # Docker provider handles this
+  plex: hp         # File provider handles this
+```
+
+### Middleware in Separate File
+
+Define middlewares outside Compose Farm's generated file:
+
+```yaml
+# /opt/traefik/dynamic.d/middlewares.yml
+http:
+  middlewares:
+    redirect-https:
+      redirectScheme:
+        scheme: https
+```
+
+## Backup Strategy
+
+### What to Backup
+
+| Item | Location | Method |
+|------|----------|--------|
+| Compose Farm config | `~/.config/compose-farm/` | Git or copy |
+| Compose files | `/opt/compose/` | Git |
+| State file | `~/.config/compose-farm/compose-farm-state.yaml` | Optional (can refresh) |
+| App data | `/opt/appdata/` | Backup solution |
+
+### Disaster Recovery
+
+```bash
+# Restore config
+cp backup/compose-farm.yaml ~/.config/compose-farm/
+
+# Refresh state from running containers
+cf refresh
+
+# Or start fresh
+cf apply
+```
+
+## Troubleshooting
+
+### Common Issues
+
+**Stack won't start:**
+```bash
+cf check myservice      # Verify mounts/networks
+cf logs myservice       # Check container logs
+```
+
+**Migration fails:**
+```bash
+cf check myservice      # Verify new host is ready
+cf init-network newhost # Create network if missing
+```
+
+**State out of sync:**
+```bash
+cf refresh --dry-run    # See differences
+cf refresh              # Sync state
+```
+
+**SSH issues:**
+```bash
+cf ssh status           # Check key status
+cf ssh setup            # Re-setup keys
+```
+
+## Security Considerations
+
+### SSH Keys
+
+- Use dedicated SSH key for Compose Farm
+- Limit key to specific hosts if possible
+- Don't store keys in Docker images
+
+### Network Exposure
+
+- Published ports are accessible from network
+- Use firewalls for sensitive services
+- Consider VPN for cross-host communication
+
+### Secrets
+
+- Don't commit `.env` files with secrets
+- Use Docker secrets or external secret management
+- Avoid secrets in compose file labels
+
+## Comparison: When to Use Alternatives
+
+| Scenario | Solution |
+|----------|----------|
+| 2-10 hosts, static stacks | **Compose Farm** |
+| Cross-host container networking | Docker Swarm |
+| Auto-scaling, self-healing | Kubernetes |
+| Infrastructure as code | Ansible + Compose Farm |
+| High availability requirements | Kubernetes or Swarm |

docs/commands.md

@@ -0,0 +1,842 @@
+---
+icon: lucide/terminal
+---
+
+# Commands Reference
+
+The Compose Farm CLI is available as both `compose-farm` and the shorter alias `cf`.
+
+## Command Overview
+
+Commands are either **Docker Compose wrappers** (`up`, `down`, `stop`, `restart`, `pull`, `logs`, `ps`, `compose`) with multi-host superpowers, or **Compose Farm originals** (`apply`, `update`, `refresh`, `check`) for orchestration Docker Compose can't do.
+
+| Category | Command | Description |
+|----------|---------|-------------|
+| **Lifecycle** | `apply` | Make reality match config |
+| | `up` | Start stacks |
+| | `down` | Stop stacks |
+| | `stop` | Stop services without removing containers |
+| | `restart` | Restart running containers |
+| | `update` | Shorthand for `up --pull --build` |
+| | `pull` | Pull latest images |
+| | `compose` | Run any docker compose command |
+| **Monitoring** | `ps` | Show stack status |
+| | `logs` | Show stack logs |
+| | `stats` | Show overview statistics |
+| | `list` | List stacks and hosts |
+| **Configuration** | `check` | Validate config and mounts |
+| | `refresh` | Sync state from reality |
+| | `init-network` | Create Docker network |
+| | `traefik-file` | Generate Traefik config |
+| | `config` | Manage config files |
+| | `ssh` | Manage SSH keys |
+| **Server** | `web` | Start web UI |
+
+## Global Options
+
+```bash
+cf --version, -v    # Show version
+cf --help, -h       # Show help
+```
+
+## Command Aliases
+
+Short aliases for frequently used commands:
+
+| Alias | Command | Alias | Command |
+|-------|---------|-------|---------|
+| `cf a` | `apply` | `cf s` | `stats` |
+| `cf l` | `logs` | `cf ls` | `list` |
+| `cf r` | `restart` | `cf rf` | `refresh` |
+| `cf u` | `update` | `cf ck` | `check` |
+| `cf p` | `pull` | `cf tf` | `traefik-file` |
+| `cf c` | `compose` | | |
+
+---
+
+## Lifecycle Commands
+
+### cf apply
+
+Make reality match your configuration. The primary reconciliation command.
+
+<video autoplay loop muted playsinline>
+  <source src="/assets/apply.webm" type="video/webm">
+</video>
+
+```bash
+cf apply [OPTIONS]
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--dry-run, -n` | Preview changes without executing |
+| `--no-orphans` | Skip stopping orphaned stacks |
+| `--no-strays` | Skip stopping stray stacks (running on wrong host) |
+| `--full, -f` | Also run up on all stacks (applies compose/env changes, triggers migrations) |
+| `--config, -c PATH` | Path to config file |
+
+**What it does:**
+
+1. Stops orphaned stacks (in state but removed from config)
+2. Stops stray stacks (running on unauthorized hosts)
+3. Migrates stacks on wrong host
+4. Starts missing stacks (in config but not running)
+
+**Examples:**
+
+```bash
+# Preview what would change
+cf apply --dry-run
+
+# Apply all changes
+cf apply
+
+# Only start/migrate, don't stop orphans
+cf apply --no-orphans
+
+# Don't stop stray stacks
+cf apply --no-strays
+
+# Also run up on all stacks (applies compose/env changes, triggers migrations)
+cf apply --full
+```
+
+---
+
+### cf up
+
+Start stacks. Auto-migrates if host assignment changed.
+
+```bash
+cf up [OPTIONS] [STACKS]...
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--all, -a` | Start all stacks |
+| `--host, -H TEXT` | Filter to stacks on this host |
+| `--service, -s TEXT` | Target a specific service within the stack |
+| `--pull` | Pull images before starting (`--pull always`) |
+| `--build` | Build images before starting |
+| `--config, -c PATH` | Path to config file |
+
+**Examples:**
+
+```bash
+# Start specific stacks
+cf up plex grafana
+
+# Start all stacks
+cf up --all
+
+# Start all stacks on a specific host
+cf up --all --host nuc
+
+# Start a specific service within a stack
+cf up immich --service database
+```
+
+**Auto-migration:**
+
+If you change a stack's host in config and run `cf up`:
+
+1. Verifies mounts/networks exist on new host
+2. Runs `down` on old host
+3. Runs `up -d` on new host
+4. Updates state
+
+---
+
+### cf down
+
+Stop stacks.
+
+```bash
+cf down [OPTIONS] [STACKS]...
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--all, -a` | Stop all stacks |
+| `--orphaned` | Stop orphaned stacks only |
+| `--host, -H TEXT` | Filter to stacks on this host |
+| `--config, -c PATH` | Path to config file |
+
+**Examples:**
+
+```bash
+# Stop specific stacks
+cf down plex
+
+# Stop all stacks
+cf down --all
+
+# Stop stacks removed from config
+cf down --orphaned
+
+# Stop all stacks on a host
+cf down --all --host nuc
+```
+
+---
+
+### cf stop
+
+Stop services without removing containers.
+
+```bash
+cf stop [OPTIONS] [STACKS]...
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--all, -a` | Stop all stacks |
+| `--service, -s TEXT` | Target a specific service within the stack |
+| `--config, -c PATH` | Path to config file |
+
+**Examples:**
+
+```bash
+# Stop specific stacks
+cf stop plex
+
+# Stop all stacks
+cf stop --all
+
+# Stop a specific service within a stack
+cf stop immich --service database
+```
+
+---
+
+### cf restart
+
+Restart running containers (`docker compose restart`). With `--service`, restarts just that service.
+
+```bash
+cf restart [OPTIONS] [STACKS]...
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--all, -a` | Restart all stacks |
+| `--service, -s TEXT` | Target a specific service within the stack |
+| `--config, -c PATH` | Path to config file |
+
+**Examples:**
+
+```bash
+cf restart plex
+cf restart --all
+
+# Restart a specific service
+cf restart immich --service database
+```
+
+---
+
+### cf update
+
+Update stacks (pull + build + up). Shorthand for `up --pull --build`. With `--service`, updates just that service.
+
+<video autoplay loop muted playsinline>
+  <source src="/assets/update.webm" type="video/webm">
+</video>
+
+```bash
+cf update [OPTIONS] [STACKS]...
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--all, -a` | Update all stacks |
+| `--service, -s TEXT` | Target a specific service within the stack |
+| `--config, -c PATH` | Path to config file |
+
+**Examples:**
+
+```bash
+# Update specific stack
+cf update plex
+
+# Update all stacks
+cf update --all
+
+# Update a specific service
+cf update immich --service database
+```
+
+---
+
+### cf pull
+
+Pull latest images.
+
+```bash
+cf pull [OPTIONS] [STACKS]...
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--all, -a` | Pull for all stacks |
+| `--service, -s TEXT` | Target a specific service within the stack |
+| `--config, -c PATH` | Path to config file |
+
+**Examples:**
+
+```bash
+cf pull plex
+cf pull --all
+
+# Pull a specific service
+cf pull immich --service database
+```
+
+---
+
+### cf compose
+
+Run any docker compose command on a stack. This is a passthrough to docker compose for commands not wrapped by cf.
+
+<video autoplay loop muted playsinline>
+  <source src="/assets/compose.webm" type="video/webm">
+</video>
+
+```bash
+cf compose [OPTIONS] STACK COMMAND [ARGS]...
+```
+
+**Arguments:**
+
+| Argument | Description |
+|----------|-------------|
+| `STACK` | Stack to operate on (use `.` for current dir) |
+| `COMMAND` | Docker compose command to run |
+| `ARGS` | Additional arguments passed to docker compose |
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--host, -H TEXT` | Filter to stacks on this host (required for multi-host stacks) |
+| `--config, -c PATH` | Path to config file |
+
+**Examples:**
+
+```bash
+# Show docker compose help
+cf compose mystack --help
+
+# View running processes
+cf compose mystack top
+
+# List images
+cf compose mystack images
+
+# Interactive shell
+cf compose mystack exec web bash
+
+# View parsed config
+cf compose mystack config
+
+# Use current directory as stack
+cf compose . ps
+```
+
+---
+
+## Monitoring Commands
+
+### cf ps
+
+Show status of stacks.
+
+```bash
+cf ps [OPTIONS] [STACKS]...
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--all, -a` | Show all stacks (default) |
+| `--host, -H TEXT` | Filter to stacks on this host |
+| `--service, -s TEXT` | Target a specific service within the stack |
+| `--config, -c PATH` | Path to config file |
+
+**Examples:**
+
+```bash
+# Show all stacks
+cf ps
+
+# Show specific stacks
+cf ps plex grafana
+
+# Filter by host
+cf ps --host nuc
+
+# Show status of a specific service
+cf ps immich --service database
+```
+
+---
+
+### cf logs
+
+Show stack logs.
+
+<video autoplay loop muted playsinline>
+  <source src="/assets/logs.webm" type="video/webm">
+</video>
+
+```bash
+cf logs [OPTIONS] [STACKS]...
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--all, -a` | Show logs for all stacks |
+| `--host, -H TEXT` | Filter to stacks on this host |
+| `--service, -s TEXT` | Target a specific service within the stack |
+| `--follow, -f` | Follow logs (live stream) |
+| `--tail, -n INTEGER` | Number of lines (default: 20 for --all, 100 otherwise) |
+| `--config, -c PATH` | Path to config file |
+
+**Examples:**
+
+```bash
+# Show last 100 lines
+cf logs plex
+
+# Follow logs
+cf logs -f plex
+
+# Show last 50 lines of multiple stacks
+cf logs -n 50 plex grafana
+
+# Show last 20 lines of all stacks
+cf logs --all
+
+# Show logs for a specific service
+cf logs immich --service database
+```
+
+---
+
+### cf stats
+
+Show overview statistics.
+
+```bash
+cf stats [OPTIONS]
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--live, -l` | Query Docker for live container counts |
+| `--config, -c PATH` | Path to config file |
+
+**Examples:**
+
+```bash
+# Config/state overview
+cf stats
+
+# Include live container counts
+cf stats --live
+```
+
+---
+
+### cf list
+
+List all stacks and their assigned hosts.
+
+```bash
+cf list [OPTIONS]
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--host, -H TEXT` | Filter to stacks on this host |
+| `--simple, -s` | Plain output for scripting (one stack per line) |
+| `--config, -c PATH` | Path to config file |
+
+**Examples:**
+
+```bash
+# List all stacks
+cf list
+
+# Filter by host
+cf list --host nas
+
+# Plain output for scripting
+cf list --simple
+
+# Combine: list stack names on a specific host
+cf list --host nuc --simple
+```
+
+---
+
+## Configuration Commands
+
+### cf check
+
+Validate configuration, mounts, and networks.
+
+```bash
+cf check [OPTIONS] [STACKS]...
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--local` | Skip SSH-based checks (faster) |
+| `--config, -c PATH` | Path to config file |
+
+**Examples:**
+
+```bash
+# Full validation with SSH
+cf check
+
+# Fast local-only validation
+cf check --local
+
+# Check specific stack and show host compatibility
+cf check jellyfin
+```
+
+---
+
+### cf refresh
+
+Update local state from running stacks.
+
+```bash
+cf refresh [OPTIONS] [STACKS]...
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--all, -a` | Refresh all stacks |
+| `--dry-run, -n` | Show what would change |
+| `--log-path, -l PATH` | Path to Dockerfarm TOML log |
+| `--config, -c PATH` | Path to config file |
+
+Without arguments, refreshes all stacks (same as `--all`). With stack names, refreshes only those stacks.
+
+**Examples:**
+
+```bash
+# Sync state with reality (all stacks)
+cf refresh
+
+# Preview changes
+cf refresh --dry-run
+
+# Refresh specific stacks only
+cf refresh plex sonarr
+```
+
+---
+
+### cf init-network
+
+Create Docker network on hosts with consistent settings.
+
+```bash
+cf init-network [OPTIONS] [HOSTS]...
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--network, -n TEXT` | Network name (default: mynetwork) |
+| `--subnet, -s TEXT` | Network subnet (default: 172.20.0.0/16) |
+| `--gateway, -g TEXT` | Network gateway (default: 172.20.0.1) |
+| `--config, -c PATH` | Path to config file |
+
+**Examples:**
+
+```bash
+# Create on all hosts
+cf init-network
+
+# Create on specific hosts
+cf init-network nuc hp
+
+# Custom network settings
+cf init-network -n production -s 10.0.0.0/16 -g 10.0.0.1
+```
+
+---
+
+### cf traefik-file
+
+Generate Traefik file-provider config from compose labels.
+
+```bash
+cf traefik-file [OPTIONS] [STACKS]...
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--all, -a` | Generate for all stacks |
+| `--output, -o PATH` | Output file (stdout if omitted) |
+| `--config, -c PATH` | Path to config file |
+
+**Examples:**
+
+```bash
+# Preview to stdout
+cf traefik-file --all
+
+# Write to file
+cf traefik-file --all -o /opt/traefik/dynamic.d/cf.yml
+
+# Specific stacks
+cf traefik-file plex jellyfin -o /opt/traefik/cf.yml
+```
+
+---
+
+### cf config
+
+Manage configuration files.
+
+```bash
+cf config COMMAND
+```
+
+**Subcommands:**
+
+| Command | Description |
+|---------|-------------|
+| `init` | Create new config with examples |
+| `init-env` | Generate .env file for Docker deployment |
+| `show` | Display config with highlighting |
+| `path` | Print config file path |
+| `validate` | Validate syntax and schema |
+| `edit` | Open in $EDITOR |
+| `symlink` | Create symlink from default location |
+
+**Options by subcommand:**
+
+| Subcommand | Options |
+|------------|---------|
+| `init` | `--path/-p PATH`, `--force/-f` |
+| `init-env` | `--path/-p PATH`, `--output/-o PATH`, `--force/-f` |
+| `show` | `--path/-p PATH`, `--raw/-r` |
+| `edit` | `--path/-p PATH` |
+| `path` | `--path/-p PATH` |
+| `validate` | `--path/-p PATH` |
+| `symlink` | `--force/-f` |
+
+**Examples:**
+
+```bash
+# Create config at default location
+cf config init
+
+# Create config at custom path
+cf config init --path /opt/compose-farm/config.yaml
+
+# Show config with syntax highlighting
+cf config show
+
+# Show raw config (for copy-paste)
+cf config show --raw
+
+# Validate config
+cf config validate
+
+# Edit config in $EDITOR
+cf config edit
+
+# Print config path
+cf config path
+
+# Create symlink to local config
+cf config symlink
+
+# Create symlink to specific file
+cf config symlink /opt/compose-farm/config.yaml
+
+# Generate .env file in current directory
+cf config init-env
+
+# Generate .env at specific path
+cf config init-env -o /opt/stacks/.env
+```
+
+---
+
+### cf ssh
+
+Manage SSH keys for passwordless authentication.
+
+```bash
+cf ssh COMMAND
+```
+
+**Subcommands:**
+
+| Command | Description |
+|---------|-------------|
+| `setup` | Generate key and copy to all hosts |
+| `status` | Show SSH key status and host connectivity |
+| `keygen` | Generate key without distributing |
+
+**Options for `cf ssh setup`:**
+
+| Option | Description |
+|--------|-------------|
+| `--config, -c PATH` | Path to config file |
+| `--force, -f` | Regenerate key even if it exists |
+
+**Options for `cf ssh status`:**
+
+| Option | Description |
+|--------|-------------|
+| `--config, -c PATH` | Path to config file |
+
+**Options for `cf ssh keygen`:**
+
+| Option | Description |
+|--------|-------------|
+| `--force, -f` | Regenerate key even if it exists |
+
+**Examples:**
+
+```bash
+# Set up SSH keys (generates and distributes)
+cf ssh setup
+
+# Check status and connectivity
+cf ssh status
+
+# Generate key only (don't distribute)
+cf ssh keygen
+```
+
+---
+
+## Server Commands
+
+### cf web
+
+Start the web UI server.
+
+```bash
+cf web [OPTIONS]
+```
+
+**Options:**
+
+| Option | Description |
+|--------|-------------|
+| `--host, -H TEXT` | Host to bind to (default: 0.0.0.0) |
+| `--port, -p INTEGER` | Port to listen on (default: 8000) |
+| `--reload, -r` | Enable auto-reload for development |
+
+**Note:** Requires web dependencies: `pip install compose-farm[web]`
+
+**Examples:**
+
+```bash
+# Start on default port
+cf web
+
+# Start on custom port
+cf web --port 3000
+
+# Development mode with auto-reload
+cf web --reload
+```
+
+---
+
+## Common Patterns
+
+### Daily Operations
+
+```bash
+# Morning: check status
+cf ps
+cf stats --live
+
+# Update a specific stack
+cf update plex
+
+# View logs
+cf logs -f plex
+```
+
+### Maintenance
+
+```bash
+# Update all stacks
+cf update --all
+
+# Refresh state after manual changes
+cf refresh
+```
+
+### Migration
+
+```bash
+# Preview what would change
+cf apply --dry-run
+
+# Move a stack: edit config, then
+cf up plex  # auto-migrates
+
+# Or reconcile everything
+cf apply
+```
+
+### Troubleshooting
+
+```bash
+# Validate config
+cf check --local
+cf check
+
+# Check specific stack
+cf check jellyfin
+
+# Sync state
+cf refresh --dry-run
+cf refresh
+```

docs/configuration.md

@@ -0,0 +1,447 @@
+---
+icon: lucide/settings
+---
+
+# Configuration Reference
+
+Compose Farm uses a YAML configuration file to define hosts and stack assignments.
+
+## Config File Location
+
+Compose Farm looks for configuration in this order:
+
+1. `-c` / `--config` flag (if provided)
+2. `CF_CONFIG` environment variable
+3. `./compose-farm.yaml` (current directory)
+4. `$XDG_CONFIG_HOME/compose-farm/compose-farm.yaml` (defaults to `~/.config`)
+
+Use `-c` / `--config` to specify a custom path:
+
+```bash
+cf ps -c /path/to/config.yaml
+```
+
+Or set the environment variable:
+
+```bash
+export CF_CONFIG=/path/to/config.yaml
+```
+
+## Examples
+
+### Single host (local-only)
+
+```yaml
+# Required: directory containing compose files
+compose_dir: /opt/stacks
+
+# Define local host
+hosts:
+  local: localhost
+
+# Map stacks to the local host
+stacks:
+  plex: local
+  grafana: local
+  nextcloud: local
+```
+
+### Multi-host (full example)
+
+```yaml
+# Required: directory containing compose files (same path on all hosts)
+compose_dir: /opt/compose
+
+# Optional: auto-regenerate Traefik config
+traefik_file: /opt/traefik/dynamic.d/compose-farm.yml
+traefik_stack: traefik
+
+# Define Docker hosts
+hosts:
+  nuc:
+    address: 192.168.1.10
+    user: docker
+  hp:
+    address: 192.168.1.11
+    user: admin
+
+# Map stacks to hosts
+stacks:
+  # Single-host stacks
+  plex: nuc
+  grafana: nuc
+  nextcloud: hp
+
+  # Multi-host stacks
+  dozzle: all                    # Run on ALL hosts
+  node-exporter: [nuc, hp]       # Run on specific hosts
+```
+
+## Settings Reference
+
+### compose_dir (required)
+
+Directory containing your compose stack folders. Must be the same path on all hosts.
+
+```yaml
+compose_dir: /opt/compose
+```
+
+**Directory structure:**
+
+```
+/opt/compose/
+├── plex/
+│   ├── docker-compose.yml    # or compose.yaml
+│   └── .env                  # optional environment file
+├── grafana/
+│   └── docker-compose.yml
+└── ...
+```
+
+Supported compose file names (checked in order):
+- `compose.yaml`
+- `compose.yml`
+- `docker-compose.yml`
+- `docker-compose.yaml`
+
+### traefik_file
+
+Path to auto-generated Traefik file-provider config. When set, Compose Farm regenerates this file after `up`, `down`, and `update` commands.
+
+```yaml
+traefik_file: /opt/traefik/dynamic.d/compose-farm.yml
+```
+
+### traefik_stack
+
+Stack name running Traefik. Stacks on the same host are skipped in file-provider config (Traefik's docker provider handles them).
+
+```yaml
+traefik_stack: traefik
+```
+
+### glances_stack
+
+Stack name running [Glances](https://nicolargo.github.io/glances/) for host resource monitoring. When set, the CLI (`cf stats --containers`) and web UI display CPU, memory, and container stats for all hosts.
+
+```yaml
+glances_stack: glances
+```
+
+The Glances stack should run on all hosts and expose port 61208. See the README for full setup instructions.
+
+## Hosts Configuration
+
+### Basic Host
+
+```yaml
+hosts:
+  myserver:
+    address: 192.168.1.10
+```
+
+### With SSH User
+
+```yaml
+hosts:
+  myserver:
+    address: 192.168.1.10
+    user: docker
+```
+
+If `user` is omitted, the current user is used.
+
+### With Custom SSH Port
+
+```yaml
+hosts:
+  myserver:
+    address: 192.168.1.10
+    user: docker
+    port: 2222  # SSH port (default: 22)
+```
+
+### Localhost
+
+For stacks running on the same machine where you invoke Compose Farm:
+
+```yaml
+hosts:
+  local: localhost
+```
+
+No SSH is used for localhost stacks.
+
+### Multiple Hosts
+
+```yaml
+hosts:
+  nuc:
+    address: 192.168.1.10
+    user: docker
+  hp:
+    address: 192.168.1.11
+    user: admin
+  truenas:
+    address: 192.168.1.100
+  local: localhost
+```
+
+## Stacks Configuration
+
+### Single-Host Stack
+
+```yaml
+stacks:
+  plex: nuc
+  grafana: nuc
+  nextcloud: hp
+```
+
+### Multi-Host Stack
+
+For stacks that need to run on every host (e.g., log shippers, monitoring agents):
+
+```yaml
+stacks:
+  # Run on ALL configured hosts
+  dozzle: all
+  promtail: all
+
+  # Run on specific hosts
+  node-exporter: [nuc, hp, truenas]
+```
+
+**Common multi-host stacks:**
+- **Dozzle** - Docker log viewer (needs local socket)
+- **Promtail/Alloy** - Log shipping (needs local socket)
+- **node-exporter** - Host metrics (needs /proc, /sys)
+- **AutoKuma** - Uptime Kuma monitors (needs local socket)
+
+### Stack Names
+
+Stack names must match directory names in `compose_dir`:
+
+```yaml
+compose_dir: /opt/compose
+stacks:
+  plex: nuc      # expects /opt/compose/plex/docker-compose.yml
+  my-app: hp     # expects /opt/compose/my-app/docker-compose.yml
+```
+
+## State File
+
+Compose Farm tracks deployment state in `compose-farm-state.yaml`, stored alongside the config file.
+
+For example, if your config is at `~/.config/compose-farm/compose-farm.yaml`, the state file will be at `~/.config/compose-farm/compose-farm-state.yaml`.
+
+```yaml
+deployed:
+  plex: nuc
+  grafana: nuc
+```
+
+This file records which stacks are deployed and on which host.
+
+**Don't edit manually.** Use `cf refresh` to sync state with reality.
+
+## Environment Variables
+
+### In Compose Files
+
+Your compose files can use `.env` files as usual:
+
+```
+/opt/compose/plex/
+├── docker-compose.yml
+└── .env
+```
+
+Compose Farm runs `docker compose` which handles `.env` automatically.
+
+### In Traefik Labels
+
+When generating Traefik config, Compose Farm resolves `${VAR}` and `${VAR:-default}` from:
+
+1. The stack's `.env` file
+2. Current environment
+
+### Compose Farm Environment Variables
+
+These environment variables configure Compose Farm itself:
+
+| Variable | Description |
+|----------|-------------|
+| `CF_CONFIG` | Path to config file |
+| `CF_WEB_STACK` | Web UI stack name (Docker only, enables self-update detection and local host inference) |
+
+**Docker deployment variables** (used in docker-compose.yml):
+
+| Variable | Description | Generated by |
+|----------|-------------|--------------|
+| `CF_COMPOSE_DIR` | Compose files directory | `cf config init-env` |
+| `CF_UID` / `CF_GID` | User/group ID for containers | `cf config init-env` |
+| `CF_HOME` / `CF_USER` | Home directory and username | `cf config init-env` |
+| `CF_SSH_DIR` | SSH keys volume mount | Manual |
+| `CF_XDG_CONFIG` | Config backup volume mount | Manual |
+
+## Config Commands
+
+### Initialize Config
+
+```bash
+cf config init
+```
+
+Creates a new config file with documented examples.
+
+### Validate Config
+
+```bash
+cf config validate
+```
+
+Checks syntax and schema.
+
+### Show Config
+
+```bash
+cf config show
+```
+
+Displays current config with syntax highlighting.
+
+### Edit Config
+
+```bash
+cf config edit
+```
+
+Opens config in `$EDITOR`.
+
+### Show Config Path
+
+```bash
+cf config path
+```
+
+Prints the config file location (useful for scripting).
+
+### Create Symlink
+
+```bash
+cf config symlink                          # Link to ./compose-farm.yaml
+cf config symlink /path/to/my-config.yaml  # Link to specific file
+```
+
+Creates a symlink from the default location (`~/.config/compose-farm/compose-farm.yaml`) to your config file. Use `--force` to overwrite an existing symlink.
+
+## Validation
+
+### Local Validation
+
+Fast validation without SSH:
+
+```bash
+cf check --local
+```
+
+Checks:
+- Config syntax
+- Stack-to-host mappings
+- Compose file existence
+
+### Full Validation
+
+```bash
+cf check
+```
+
+Additional SSH-based checks:
+- Host connectivity
+- Mount point existence
+- Docker network existence
+- Traefik label validation
+
+### Stack-Specific Check
+
+```bash
+cf check jellyfin
+```
+
+Shows which hosts can run the stack (have required mounts/networks).
+
+## Example Configurations
+
+### Minimal
+
+```yaml
+compose_dir: /opt/compose
+
+hosts:
+  server: 192.168.1.10
+
+stacks:
+  myapp: server
+```
+
+### Home Lab
+
+```yaml
+compose_dir: /opt/compose
+
+hosts:
+  nuc:
+    address: 192.168.1.10
+    user: docker
+  nas:
+    address: 192.168.1.100
+    user: admin
+
+stacks:
+  # Media
+  plex: nuc
+  jellyfin: nuc
+  immich: nuc
+
+  # Infrastructure
+  traefik: nuc
+  portainer: nuc
+
+  # Monitoring (on all hosts)
+  dozzle: all
+```
+
+### Production
+
+```yaml
+compose_dir: /opt/compose
+traefik_file: /opt/traefik/dynamic.d/cf.yml
+traefik_stack: traefik
+
+hosts:
+  web-1:
+    address: 10.0.1.10
+    user: deploy
+  web-2:
+    address: 10.0.1.11
+    user: deploy
+  db:
+    address: 10.0.1.20
+    user: deploy
+
+stacks:
+  # Load balanced
+  api: [web-1, web-2]
+
+  # Single instance
+  postgres: db
+  redis: db
+
+  # Infrastructure
+  traefik: web-1
+
+  # Monitoring
+  promtail: all
+```

docs/demos/README.md

@@ -0,0 +1,17 @@
+# Demo Recordings
+
+Demo recording infrastructure for Compose Farm documentation.
+
+## Structure
+
+```
+docs/demos/
+├── cli/        # VHS-based CLI terminal recordings
+└── web/        # Playwright-based web UI recordings
+```
+
+## Output
+
+All recordings output to `docs/assets/` as WebM (primary) and GIF (fallback).
+
+See subdirectory READMEs for usage.

docs/demos/cli/README.md

@@ -0,0 +1,33 @@
+# CLI Demo Recordings
+
+VHS-based terminal demo recordings for Compose Farm CLI.
+
+## Requirements
+
+- [VHS](https://github.com/charmbracelet/vhs): `go install github.com/charmbracelet/vhs@latest`
+
+## Usage
+
+```bash
+# Record all demos
+python docs/demos/cli/record.py
+
+# Record specific demos
+python docs/demos/cli/record.py quickstart migration
+```
+
+## Demos
+
+| Tape | Description |
+|------|-------------|
+| `install.tape` | Installing with `uv tool install` |
+| `quickstart.tape` | `cf ps`, `cf up`, `cf logs` |
+| `logs.tape` | Viewing logs |
+| `compose.tape` | `cf compose` passthrough (--help, images, exec) |
+| `update.tape` | `cf update` |
+| `migration.tape` | Service migration |
+| `apply.tape` | `cf apply` |
+
+## Output
+
+GIF and WebM files saved to `docs/assets/`.

docs/demos/cli/apply.tape

@@ -0,0 +1,39 @@
+# Apply Demo
+# Shows cf apply previewing and reconciling state
+
+Output docs/assets/apply.gif
+Output docs/assets/apply.webm
+
+Set Shell "bash"
+Set FontSize 14
+Set Width 900
+Set Height 600
+Set Theme "Catppuccin Mocha"
+Set TypingSpeed 50ms
+
+Type "# Preview what would change"
+Enter
+Sleep 500ms
+
+Type "cf apply --dry-run"
+Enter
+Wait
+
+Type "# Check current status"
+Enter
+Sleep 500ms
+
+Type "cf stats"
+Enter
+Wait+Screen /Summary/
+Sleep 2s
+
+Type "# Apply the changes"
+Enter
+Sleep 500ms
+
+Type "cf apply"
+Enter
+# Wait for shell prompt (command complete)
+Wait
+Sleep 4s

docs/demos/cli/compose.tape

@@ -0,0 +1,50 @@
+# Compose Demo
+# Shows that cf compose passes through ANY docker compose command
+
+Output docs/assets/compose.gif
+Output docs/assets/compose.webm
+
+Set Shell "bash"
+Set FontSize 14
+Set Width 900
+Set Height 550
+Set Theme "Catppuccin Mocha"
+Set TypingSpeed 50ms
+
+Type "# cf compose runs ANY docker compose command on the right host"
+Enter
+Sleep 500ms
+
+Type "# See ALL available compose commands"
+Enter
+Sleep 500ms
+
+Type "cf compose immich --help"
+Enter
+Sleep 4s
+
+Type "# Show images"
+Enter
+Sleep 500ms
+
+Type "cf compose immich images"
+Enter
+Wait+Screen /immich/
+Sleep 2s
+
+Type "# Open shell in a container"
+Enter
+Sleep 500ms
+
+Type "cf compose immich exec immich-machine-learning sh"
+Enter
+Wait+Screen /#/
+Sleep 1s
+
+Type "python3 --version"
+Enter
+Sleep 1s
+
+Type "exit"
+Enter
+Sleep 500ms

docs/demos/cli/install.tape

@@ -0,0 +1,42 @@
+# Installation Demo
+# Shows installing compose-farm with uv
+
+Output docs/assets/install.gif
+Output docs/assets/install.webm
+
+Set Shell "bash"
+Set FontSize 14
+Set Width 900
+Set Height 600
+Set Theme "Catppuccin Mocha"
+Set TypingSpeed 50ms
+Env FORCE_COLOR "1"
+
+Hide
+Type "export PATH=$HOME/.local/bin:$PATH && uv tool uninstall compose-farm 2>/dev/null; clear"
+Enter
+Show
+Type "# Install with uv (recommended)"
+Enter
+Sleep 500ms
+
+Type "uv tool install compose-farm"
+Enter
+Wait+Screen /Installed|already installed/
+
+Type "# Verify installation"
+Enter
+Sleep 500ms
+
+Type "cf --version"
+Enter
+Wait+Screen /compose-farm/
+Sleep 1s
+
+Type "cf --help | less"
+Enter
+Sleep 2s
+PageDown
+Sleep 2s
+Type "q"
+Sleep 2s

docs/demos/cli/logs.tape

@@ -0,0 +1,21 @@
+# Logs Demo
+# Shows viewing stack logs
+
+Output docs/assets/logs.gif
+Output docs/assets/logs.webm
+
+Set Shell "bash"
+Set FontSize 14
+Set Width 900
+Set Height 550
+Set Theme "Catppuccin Mocha"
+Set TypingSpeed 50ms
+
+Type "# View recent logs"
+Enter
+Sleep 500ms
+
+Type "cf logs immich --tail 20"
+Enter
+Wait+Screen /immich/
+Sleep 2s

docs/demos/cli/migration.tape

@@ -0,0 +1,71 @@
+# Migration Demo
+# Shows automatic stack migration when host changes
+
+Output docs/assets/migration.gif
+Output docs/assets/migration.webm
+
+Set Shell "bash"
+Set FontSize 14
+Set Width 1000
+Set Height 600
+Set Theme "Catppuccin Mocha"
+Set TypingSpeed 50ms
+
+Type "# Current status: audiobookshelf on 'nas'"
+Enter
+Sleep 500ms
+
+Type "cf ps audiobookshelf"
+Enter
+Wait+Screen /PORTS/
+
+Type "# Edit config to move it to 'anton'"
+Enter
+Sleep 1s
+
+Type "nvim /opt/stacks/compose-farm.yaml"
+Enter
+Wait+Screen /stacks:/
+
+# Search for audiobookshelf
+Type "/audiobookshelf"
+Enter
+Sleep 1s
+
+# Move to the host value (nas) and change it
+Type "f:"
+Sleep 500ms
+Type "w"
+Sleep 500ms
+Type "ciw"
+Sleep 500ms
+Type "anton"
+Escape
+Sleep 1s
+
+# Save and quit
+Type ":wq"
+Enter
+Sleep 1s
+
+Type "# Run up - automatically migrates!"
+Enter
+Sleep 500ms
+
+Type "cf up audiobookshelf"
+Enter
+# Wait for migration phases: first the stop on old host
+Wait+Screen /Migrating|down/
+# Then wait for start on new host
+Wait+Screen /Starting|up/
+# Finally wait for completion
+Wait
+
+Type "# Verify: audiobookshelf now on 'anton'"
+Enter
+Sleep 500ms
+
+Type "cf ps audiobookshelf"
+Enter
+Wait+Screen /PORTS/
+Sleep 3s

docs/demos/cli/quickstart.tape

@@ -0,0 +1,91 @@
+# Quick Start Demo
+# Shows basic cf commands
+
+Output docs/assets/quickstart.gif
+Output docs/assets/quickstart.webm
+
+Set Shell "bash"
+Set FontSize 14
+Set Width 900
+Set Height 600
+Set Theme "Catppuccin Mocha"
+Set FontFamily "FiraCode Nerd Font"
+Set TypingSpeed 50ms
+Env BAT_PAGING "always"
+
+Type "# Config is just: stack  host"
+Enter
+Sleep 500ms
+
+Type "# First, define your hosts..."
+Enter
+Sleep 500ms
+
+Type "bat -r 1:16 compose-farm.yaml"
+Enter
+Sleep 3s
+Type "q"
+Sleep 500ms
+
+Type "# Then map each stack to a host"
+Enter
+Sleep 500ms
+
+Type "bat -r 17:35 compose-farm.yaml"
+Enter
+Sleep 3s
+Type "q"
+Sleep 500ms
+
+Type "# Check stack status"
+Enter
+Sleep 500ms
+
+Type "cf ps immich"
+Enter
+Wait+Screen /PORTS/
+
+Type "# Start a stack"
+Enter
+Sleep 500ms
+
+Type "cf up immich"
+Enter
+Wait
+
+Type "# View logs"
+Enter
+Sleep 500ms
+
+Type "cf logs immich --tail 5"
+Enter
+Wait+Screen /immich/
+Sleep 2s
+
+Type "#  The magic: move between hosts (nas  anton)"
+Enter
+Sleep 500ms
+
+Type "# Change host in config (using sed)"
+Enter
+Sleep 500ms
+
+Type "sed -i 's/audiobookshelf: nas/audiobookshelf: anton/' compose-farm.yaml"
+Enter
+Sleep 500ms
+
+Type "# Apply changes - auto-migrates!"
+Enter
+Sleep 500ms
+
+Type "cf apply"
+Enter
+Sleep 15s
+
+Type "# Verify: now on anton"
+Enter
+Sleep 500ms
+
+Type "cf ps audiobookshelf"
+Enter
+Sleep 5s

docs/demos/cli/record.py

@@ -0,0 +1,134 @@
+#!/usr/bin/env python3
+"""Record CLI demos using VHS."""
+
+import shutil
+import subprocess
+import sys
+from pathlib import Path
+
+from rich.console import Console
+
+from compose_farm.config import load_config
+from compose_farm.state import load_state
+
+console = Console()
+SCRIPT_DIR = Path(__file__).parent
+STACKS_DIR = Path("/opt/stacks")
+CONFIG_FILE = STACKS_DIR / "compose-farm.yaml"
+OUTPUT_DIR = SCRIPT_DIR.parent.parent / "assets"
+
+DEMOS = ["install", "quickstart", "logs", "compose", "update", "migration", "apply"]
+
+
+def _run(cmd: list[str], **kw) -> bool:
+    return subprocess.run(cmd, check=False, **kw).returncode == 0
+
+
+def _set_config(host: str) -> None:
+    """Set audiobookshelf host in config file."""
+    _run(["sed", "-i", f"s/audiobookshelf: .*/audiobookshelf: {host}/", str(CONFIG_FILE)])
+
+
+def _get_hosts() -> tuple[str | None, str | None]:
+    """Return (config_host, state_host) for audiobookshelf."""
+    config = load_config()
+    state = load_state(config)
+    return config.stacks.get("audiobookshelf"), state.get("audiobookshelf")
+
+
+def _setup_state(demo: str) -> bool:
+    """Set up required state for demo. Returns False on failure."""
+    if demo not in ("migration", "apply"):
+        return True
+
+    config_host, state_host = _get_hosts()
+
+    if demo == "migration":
+        # Migration needs audiobookshelf on nas in BOTH config and state
+        if config_host != "nas":
+            console.print("[yellow]Setting up: config → nas[/yellow]")
+            _set_config("nas")
+        if state_host != "nas":
+            console.print("[yellow]Setting up: state → nas[/yellow]")
+            if not _run(["cf", "apply"], cwd=STACKS_DIR):
+                return False
+
+    elif demo == "apply":
+        # Apply needs config=nas, state=anton (so there's something to apply)
+        if config_host != "nas":
+            console.print("[yellow]Setting up: config → nas[/yellow]")
+            _set_config("nas")
+        if state_host == "nas":
+            console.print("[yellow]Setting up: state → anton[/yellow]")
+            _set_config("anton")
+            if not _run(["cf", "apply"], cwd=STACKS_DIR):
+                return False
+            _set_config("nas")
+
+    return True
+
+
+def _record(name: str, index: int, total: int) -> bool:
+    """Record a single demo."""
+    console.print(f"[cyan][{index}/{total}][/cyan] [green]Recording:[/green] {name}")
+    if _run(["vhs", str(SCRIPT_DIR / f"{name}.tape")], cwd=STACKS_DIR):
+        console.print("[green]  ✓ Done[/green]")
+        return True
+    console.print("[red]  ✗ Failed[/red]")
+    return False
+
+
+def _reset_after(demo: str, next_demo: str | None) -> None:
+    """Reset state after demos that modify audiobookshelf."""
+    if demo not in ("quickstart", "migration"):
+        return
+    _set_config("nas")
+    if next_demo != "apply":  # Let apply demo show the migration
+        _run(["cf", "apply"], cwd=STACKS_DIR)
+
+
+def _restore_config(original: str) -> None:
+    """Restore original config and sync state."""
+    console.print("[yellow]Restoring original config...[/yellow]")
+    CONFIG_FILE.write_text(original)
+    _run(["cf", "apply"], cwd=STACKS_DIR)
+
+
+def _main() -> int:
+    if not shutil.which("vhs"):
+        console.print("[red]VHS not found. Install: brew install vhs[/red]")
+        return 1
+
+    if not _run(["git", "-C", str(STACKS_DIR), "diff", "--quiet", "compose-farm.yaml"]):
+        console.print("[red]compose-farm.yaml has uncommitted changes[/red]")
+        return 1
+
+    demos = [d for d in sys.argv[1:] if d in DEMOS] or DEMOS
+    if sys.argv[1:] and not demos:
+        console.print(f"[red]Unknown demo. Available: {', '.join(DEMOS)}[/red]")
+        return 1
+
+    # Save original config to restore after recording
+    original_config = CONFIG_FILE.read_text()
+
+    try:
+        for i, demo in enumerate(demos, 1):
+            if not _setup_state(demo):
+                return 1
+            if not _record(demo, i, len(demos)):
+                return 1
+            _reset_after(demo, demos[i] if i < len(demos) else None)
+    finally:
+        _restore_config(original_config)
+
+    # Move outputs
+    OUTPUT_DIR.mkdir(exist_ok=True)
+    for f in (STACKS_DIR / "docs/assets").glob("*.[gw]*"):
+        shutil.move(str(f), str(OUTPUT_DIR / f.name))
+
+    console.print(f"\n[green]Done![/green] Saved to {OUTPUT_DIR}")
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(_main())

docs/demos/cli/update.tape

@@ -0,0 +1,32 @@
+# Update Demo
+# Shows updating stacks (only recreates containers if images changed)
+
+Output docs/assets/update.gif
+Output docs/assets/update.webm
+
+Set Shell "bash"
+Set FontSize 14
+Set Width 900
+Set Height 500
+Set Theme "Catppuccin Mocha"
+Set TypingSpeed 50ms
+
+Type "# Update a single stack"
+Enter
+Sleep 500ms
+
+Type "cf update grocy"
+Enter
+# Wait for command to complete (chain waits for longer timeout)
+Wait+Screen /pull/
+Wait+Screen /grocy/
+Wait@60s
+
+Type "# Check current status"
+Enter
+Sleep 500ms
+
+Type "cf ps grocy"
+Enter
+Wait+Screen /PORTS/
+Sleep 1s

docs/demos/web/README.md

@@ -0,0 +1,45 @@
+# Web UI Demo Recordings
+
+Playwright-based demo recording for Compose Farm web UI.
+
+## Requirements
+
+- Chromium: `playwright install chromium`
+- ffmpeg: `apt install ffmpeg` or `brew install ffmpeg`
+
+## Usage
+
+```bash
+# Record all demos
+python docs/demos/web/record.py
+
+# Record specific demo
+python docs/demos/web/record.py navigation
+```
+
+## Demos
+
+| Demo | Description |
+|------|-------------|
+| `navigation` | Command palette fuzzy search and navigation |
+| `stack` | Stack restart/logs via command palette |
+| `themes` | Theme switching with arrow key preview |
+| `workflow` | Full workflow: filter, navigate, logs, themes |
+| `console` | Console terminal running cf commands |
+| `shell` | Container shell exec with top |
+
+## Output
+
+WebM and GIF files saved to `docs/assets/web-{demo}.{webm,gif}`.
+
+## Files
+
+- `record.py` - Orchestration script
+- `conftest.py` - Playwright fixtures, helper functions
+- `demo_*.py` - Individual demo scripts
+
+## Notes
+
+- Uses real config at `/opt/stacks/compose-farm.yaml`
+- Adjust `pause(page, ms)` calls to control timing
+- Viewport: 1280x720

docs/demos/web/__init__.py

@@ -0,0 +1 @@
+"""Web UI demo recording scripts."""

docs/demos/web/conftest.py

@@ -0,0 +1,302 @@
+"""Shared fixtures for web UI demo recordings.
+
+Based on tests/web/test_htmx_browser.py patterns for consistency.
+"""
+
+from __future__ import annotations
+
+import os
+import re
+import shutil
+import socket
+import threading
+import time
+import urllib.request
+from pathlib import Path
+from typing import TYPE_CHECKING, Any
+from unittest.mock import patch
+
+import pytest
+import uvicorn
+
+from compose_farm.config import Config as CFConfig
+from compose_farm.config import load_config
+from compose_farm.executor import (
+    get_container_compose_labels as _original_get_compose_labels,
+)
+from compose_farm.glances import ContainerStats
+from compose_farm.glances import fetch_container_stats as _original_fetch_container_stats
+from compose_farm.state import load_state as _original_load_state
+from compose_farm.web.cdn import CDN_ASSETS, ensure_vendor_cache
+
+# NOTE: Do NOT import create_app here - it must be imported AFTER patches are applied
+# to ensure the patched get_config is used by all route modules
+
+if TYPE_CHECKING:
+    from collections.abc import Generator
+
+    from playwright.sync_api import BrowserContext, Page, Route
+
+# Substrings to exclude from demo recordings (case-insensitive)
+DEMO_EXCLUDE_PATTERNS = {"arr", "vpn", "tash"}
+
+
+def _should_exclude(name: str) -> bool:
+    """Check if a stack/container name should be excluded from demo."""
+    name_lower = name.lower()
+    return any(pattern in name_lower for pattern in DEMO_EXCLUDE_PATTERNS)
+
+
+def _get_filtered_config() -> CFConfig:
+    """Load config but filter out excluded stacks."""
+    config = load_config()
+    filtered_stacks = {
+        name: host for name, host in config.stacks.items() if not _should_exclude(name)
+    }
+    return CFConfig(
+        compose_dir=config.compose_dir,
+        hosts=config.hosts,
+        stacks=filtered_stacks,
+        traefik_file=config.traefik_file,
+        traefik_stack=config.traefik_stack,
+        glances_stack=config.glances_stack,
+        config_path=config.config_path,
+    )
+
+
+def _get_filtered_state(config: CFConfig) -> dict[str, str | list[str]]:
+    """Load state but filter out excluded stacks."""
+    state = _original_load_state(config)
+    return {name: host for name, host in state.items() if not _should_exclude(name)}
+
+
+async def _filtered_fetch_container_stats(
+    host_name: str,
+    host_address: str,
+    port: int = 61208,
+    request_timeout: float = 10.0,
+) -> tuple[list[ContainerStats] | None, str | None]:
+    """Fetch container stats but filter out excluded containers."""
+    containers, error = await _original_fetch_container_stats(
+        host_name, host_address, port, request_timeout
+    )
+    if containers:
+        # Filter by container name (stack is empty at this point)
+        containers = [c for c in containers if not _should_exclude(c.name)]
+    return containers, error
+
+
+async def _filtered_get_compose_labels(
+    config: CFConfig,
+    host_name: str,
+) -> dict[str, tuple[str, str]]:
+    """Get compose labels but filter out excluded stacks."""
+    labels = await _original_get_compose_labels(config, host_name)
+    # Filter out containers whose stack (project) name should be excluded
+    return {
+        name: (stack, service)
+        for name, (stack, service) in labels.items()
+        if not _should_exclude(stack)
+    }
+
+
+@pytest.fixture(scope="session")
+def vendor_cache(request: pytest.FixtureRequest) -> Path:
+    """Download CDN assets once and cache to disk for faster recordings."""
+    cache_dir = Path(str(request.config.rootdir)) / ".pytest_cache" / "vendor"
+    return ensure_vendor_cache(cache_dir)
+
+
+@pytest.fixture(scope="session")
+def browser_type_launch_args() -> dict[str, str]:
+    """Configure Playwright to use system Chromium if available."""
+    for name in ["chromium", "chromium-browser", "google-chrome", "chrome"]:
+        path = shutil.which(name)
+        if path:
+            return {"executable_path": path}
+    return {}
+
+
+# Path to real compose-farm config
+REAL_CONFIG_PATH = Path("/opt/stacks/compose-farm.yaml")
+
+
+@pytest.fixture(scope="module")
+def server_url() -> Generator[str, None, None]:
+    """Start demo server using real config (with filtered stacks) and return URL."""
+    os.environ["CF_CONFIG"] = str(REAL_CONFIG_PATH)
+
+    # Patch at source module level so all callers get filtered versions
+    patches = [
+        # Patch load_config at source - get_config() calls this internally
+        patch("compose_farm.config.load_config", _get_filtered_config),
+        # Patch load_state at source and where imported
+        patch("compose_farm.state.load_state", _get_filtered_state),
+        patch("compose_farm.web.routes.pages.load_state", _get_filtered_state),
+        # Patch container fetch to filter out excluded containers (Live Stats page)
+        patch("compose_farm.glances.fetch_container_stats", _filtered_fetch_container_stats),
+        # Patch compose labels to filter out excluded stacks
+        patch("compose_farm.executor.get_container_compose_labels", _filtered_get_compose_labels),
+    ]
+
+    for p in patches:
+        p.start()
+
+    # Import create_app AFTER patches are started so route modules see patched get_config
+    from compose_farm.web.app import create_app  # noqa: PLC0415
+
+    with socket.socket() as s:
+        s.bind(("127.0.0.1", 0))
+        port = s.getsockname()[1]
+
+    app = create_app()
+    uvicorn_config = uvicorn.Config(app, host="127.0.0.1", port=port, log_level="error")
+    server = uvicorn.Server(uvicorn_config)
+
+    thread = threading.Thread(target=server.run, daemon=True)
+    thread.start()
+
+    url = f"http://127.0.0.1:{port}"
+    server_ready = False
+    for _ in range(50):
+        try:
+            urllib.request.urlopen(url, timeout=0.5)  # noqa: S310
+            server_ready = True
+            break
+        except Exception:
+            time.sleep(0.1)
+
+    if not server_ready:
+        msg = f"Demo server failed to start on {url}"
+        raise RuntimeError(msg)
+
+    yield url
+
+    server.should_exit = True
+    thread.join(timeout=2)
+    os.environ.pop("CF_CONFIG", None)
+
+    for p in patches:
+        p.stop()
+
+
+@pytest.fixture(scope="module")
+def recording_output_dir(tmp_path_factory: pytest.TempPathFactory) -> Path:
+    """Directory for video recordings."""
+    return Path(tmp_path_factory.mktemp("recordings"))
+
+
+@pytest.fixture
+def recording_context(
+    browser: Any,  # pytest-playwright's browser fixture
+    vendor_cache: Path,
+    recording_output_dir: Path,
+) -> Generator[BrowserContext, None, None]:
+    """Browser context with video recording enabled."""
+    context = browser.new_context(
+        viewport={"width": 1280, "height": 720},
+        record_video_dir=str(recording_output_dir),
+        record_video_size={"width": 1280, "height": 720},
+    )
+
+    # Set up CDN interception
+    cache = {url: (vendor_cache / f, ct) for url, (f, ct) in CDN_ASSETS.items()}
+
+    def handle_cdn(route: Route) -> None:
+        url = route.request.url
+        for url_prefix, (filepath, content_type) in cache.items():
+            if url.startswith(url_prefix):
+                route.fulfill(status=200, content_type=content_type, body=filepath.read_bytes())
+                return
+        print(f"UNCACHED CDN request: {url}")
+        route.abort("failed")
+
+    context.route(re.compile(r"https://(cdn\.jsdelivr\.net|unpkg\.com)/.*"), handle_cdn)
+
+    yield context
+    context.close()
+
+
+@pytest.fixture
+def recording_page(recording_context: BrowserContext) -> Generator[Page, None, None]:
+    """Page with recording and slow motion enabled."""
+    page = recording_context.new_page()
+    yield page
+    page.close()
+
+
+@pytest.fixture
+def wide_recording_context(
+    browser: Any,  # pytest-playwright's browser fixture
+    recording_output_dir: Path,
+) -> Generator[BrowserContext, None, None]:
+    """Browser context with wider viewport for demos needing more horizontal space.
+
+    NOTE: This fixture does NOT use CDN interception (unlike recording_context).
+    CDN interception was causing inline scripts from containers.html to be
+    removed from the DOM, likely due to Tailwind's browser plugin behavior.
+    """
+    context = browser.new_context(
+        viewport={"width": 1920, "height": 1080},
+        record_video_dir=str(recording_output_dir),
+        record_video_size={"width": 1920, "height": 1080},
+    )
+
+    yield context
+    context.close()
+
+
+@pytest.fixture
+def wide_recording_page(wide_recording_context: BrowserContext) -> Generator[Page, None, None]:
+    """Page with wider viewport for demos needing more horizontal space."""
+    page = wide_recording_context.new_page()
+    yield page
+    page.close()
+
+
+# Demo helper functions
+
+
+def pause(page: Page, ms: int = 500) -> None:
+    """Pause for visibility in recording."""
+    page.wait_for_timeout(ms)
+
+
+def slow_type(page: Page, selector: str, text: str, delay: int = 100) -> None:
+    """Type with visible delay between keystrokes."""
+    page.type(selector, text, delay=delay)
+
+
+def open_command_palette(page: Page) -> None:
+    """Open command palette with Ctrl+K."""
+    page.keyboard.press("Control+k")
+    page.wait_for_selector("#cmd-palette[open]", timeout=2000)
+    pause(page, 300)
+
+
+def close_command_palette(page: Page) -> None:
+    """Close command palette with Escape."""
+    page.keyboard.press("Escape")
+    page.wait_for_selector("#cmd-palette:not([open])", timeout=2000)
+    pause(page, 200)
+
+
+def wait_for_sidebar(page: Page) -> None:
+    """Wait for sidebar to load with stacks."""
+    page.wait_for_selector("#sidebar-stacks", timeout=5000)
+    pause(page, 300)
+
+
+def navigate_to_stack(page: Page, stack: str) -> None:
+    """Navigate to a stack page via sidebar click."""
+    page.locator("#sidebar-stacks a", has_text=stack).click()
+    page.wait_for_url(f"**/stack/{stack}", timeout=5000)
+    pause(page, 500)
+
+
+def select_command(page: Page, command: str) -> None:
+    """Filter and select a command from the palette."""
+    page.locator("#cmd-input").fill(command)
+    pause(page, 300)
+    page.keyboard.press("Enter")
+    pause(page, 200)

docs/demos/web/demo_console.py

@@ -0,0 +1,77 @@
+"""Demo: Console terminal.
+
+Records a ~30 second demo showing:
+- Navigating to Console page
+- Running cf commands in the terminal
+- Showing the Compose Farm config in Monaco editor
+
+Run: pytest docs/demos/web/demo_console.py -v --no-cov
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import pytest
+from conftest import (
+    pause,
+    slow_type,
+    wait_for_sidebar,
+)
+
+if TYPE_CHECKING:
+    from playwright.sync_api import Page
+
+
+@pytest.mark.browser  # type: ignore[misc]
+def test_demo_console(recording_page: Page, server_url: str) -> None:
+    """Record console terminal demo."""
+    page = recording_page
+
+    # Start on dashboard
+    page.goto(server_url)
+    wait_for_sidebar(page)
+    pause(page, 800)
+
+    # Navigate to Console page via sidebar menu
+    page.locator(".menu a", has_text="Console").click()
+    page.wait_for_url("**/console", timeout=5000)
+    pause(page, 1000)
+
+    # Wait for terminal to be ready (auto-connects)
+    page.wait_for_selector("#console-terminal .xterm", timeout=10000)
+    pause(page, 1500)
+
+    # Run fastfetch first
+    slow_type(page, "#console-terminal .xterm-helper-textarea", "fastfetch", delay=80)
+    pause(page, 300)
+    page.keyboard.press("Enter")
+    pause(page, 2500)  # Wait for output
+
+    # Type cf stats command
+    slow_type(page, "#console-terminal .xterm-helper-textarea", "cf stats", delay=80)
+    pause(page, 300)
+    page.keyboard.press("Enter")
+    pause(page, 3000)  # Wait for output
+
+    # Type cf ps command
+    slow_type(page, "#console-terminal .xterm-helper-textarea", "cf ps grocy", delay=80)
+    pause(page, 300)
+    page.keyboard.press("Enter")
+    pause(page, 2500)  # Wait for output
+
+    # Smoothly scroll down to show the Editor section with Compose Farm config
+    page.evaluate("""
+        const editor = document.getElementById('console-editor');
+        if (editor) {
+            editor.scrollIntoView({ behavior: 'smooth', block: 'center' });
+        }
+    """)
+    pause(page, 1200)  # Wait for smooth scroll animation
+
+    # Wait for Monaco editor to load with config content
+    page.wait_for_selector("#console-editor .monaco-editor", timeout=10000)
+    pause(page, 2500)  # Let viewer see the Compose Farm config file
+
+    # Final pause
+    pause(page, 800)

docs/demos/web/demo_live_stats.py

@@ -0,0 +1,85 @@
+"""Demo: Live Stats page.
+
+Records a ~20 second demo showing:
+- Navigating to Live Stats via command palette
+- Container table with real-time stats
+- Filtering containers
+- Sorting by different columns
+- Auto-refresh countdown
+
+Run: pytest docs/demos/web/demo_live_stats.py -v --no-cov
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import pytest
+from conftest import (
+    open_command_palette,
+    pause,
+    slow_type,
+    wait_for_sidebar,
+)
+
+if TYPE_CHECKING:
+    from playwright.sync_api import Page
+
+
+@pytest.mark.browser  # type: ignore[misc]
+def test_demo_live_stats(wide_recording_page: Page, server_url: str) -> None:
+    """Record Live Stats page demo."""
+    page = wide_recording_page
+
+    # Start on dashboard
+    page.goto(server_url)
+    wait_for_sidebar(page)
+    pause(page, 1000)
+
+    # Navigate to Live Stats via command palette
+    open_command_palette(page)
+    pause(page, 400)
+    slow_type(page, "#cmd-input", "live", delay=100)
+    pause(page, 500)
+    page.keyboard.press("Enter")
+    page.wait_for_url("**/live-stats", timeout=5000)
+
+    # Wait for containers to load (may take ~10s on first load due to SSH)
+    page.wait_for_selector("#container-rows tr:not(:has(.loading))", timeout=30000)
+    pause(page, 2000)  # Let viewer see the full table with timer
+
+    # Demonstrate filtering
+    slow_type(page, "#filter-input", "grocy", delay=100)
+    pause(page, 1500)  # Show filtered results
+
+    # Clear filter
+    page.fill("#filter-input", "")
+    pause(page, 1000)
+
+    # Sort by memory (click header)
+    page.click("th:has-text('Mem')")
+    pause(page, 1500)
+
+    # Sort by CPU
+    page.click("th:has-text('CPU')")
+    pause(page, 1500)
+
+    # Sort by host
+    page.click("th:has-text('Host')")
+    pause(page, 1500)
+
+    # Watch auto-refresh timer count down
+    pause(page, 3500)  # Wait for refresh to happen
+
+    # Hover on action menu to show pause behavior
+    action_btn = page.locator('button[onclick^="openActionMenu"]').first
+    action_btn.scroll_into_view_if_needed()
+    action_btn.hover()
+    pause(page, 2000)  # Show paused state (timer shows ⏸) and action menu
+
+    # Move away to close menu and resume refresh
+    page.locator("h2").first.hover()  # Move to header
+    pause(page, 3500)  # Watch countdown resume and refresh happen
+
+    # Final pause
+    pause(page, 1000)

docs/demos/web/demo_navigation.py

@@ -0,0 +1,74 @@
+"""Demo: Command palette navigation.
+
+Records a ~15 second demo showing:
+- Opening command palette with Ctrl+K
+- Fuzzy search filtering
+- Arrow key navigation
+- Stack and page navigation
+
+Run: pytest docs/demos/web/demo_navigation.py -v --no-cov
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import pytest
+from conftest import (
+    open_command_palette,
+    pause,
+    slow_type,
+    wait_for_sidebar,
+)
+
+if TYPE_CHECKING:
+    from playwright.sync_api import Page
+
+
+@pytest.mark.browser  # type: ignore[misc]
+def test_demo_navigation(recording_page: Page, server_url: str) -> None:
+    """Record command palette navigation demo."""
+    page = recording_page
+
+    # Start on dashboard
+    page.goto(server_url)
+    wait_for_sidebar(page)
+    pause(page, 1000)  # Let viewer see dashboard
+
+    # Open command palette with keyboard shortcut
+    open_command_palette(page)
+    pause(page, 500)
+
+    # Type partial stack name for fuzzy search
+    slow_type(page, "#cmd-input", "grocy", delay=120)
+    pause(page, 800)
+
+    # Arrow down to show selection movement
+    page.keyboard.press("ArrowDown")
+    pause(page, 400)
+    page.keyboard.press("ArrowUp")
+    pause(page, 400)
+
+    # Press Enter to navigate to stack
+    page.keyboard.press("Enter")
+    page.wait_for_url("**/stack/grocy", timeout=5000)
+    pause(page, 1500)  # Show stack page
+
+    # Open palette again to navigate elsewhere
+    open_command_palette(page)
+    pause(page, 400)
+
+    # Navigate to another stack (immich) to show more navigation
+    slow_type(page, "#cmd-input", "imm", delay=120)
+    pause(page, 600)
+    page.keyboard.press("Enter")
+    page.wait_for_url("**/stack/immich", timeout=5000)
+    pause(page, 1200)  # Show immich stack page
+
+    # Open palette one more time, navigate back to dashboard
+    open_command_palette(page)
+    slow_type(page, "#cmd-input", "dashb", delay=120)
+    pause(page, 500)
+    page.keyboard.press("Enter")
+    page.wait_for_url(server_url, timeout=5000)
+    pause(page, 1000)  # Final dashboard view

docs/demos/web/demo_shell.py

@@ -0,0 +1,106 @@
+"""Demo: Container shell exec via command palette.
+
+Records a ~35 second demo showing:
+- Navigating to immich stack (multiple containers)
+- Using command palette with fuzzy matching ("sh mach") to open shell
+- Running a command
+- Using command palette to switch to server container shell
+- Running another command
+
+Run: pytest docs/demos/web/demo_shell.py -v --no-cov
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import pytest
+from conftest import (
+    open_command_palette,
+    pause,
+    slow_type,
+    wait_for_sidebar,
+)
+
+if TYPE_CHECKING:
+    from playwright.sync_api import Page
+
+
+@pytest.mark.browser  # type: ignore[misc]
+def test_demo_shell(recording_page: Page, server_url: str) -> None:
+    """Record container shell demo."""
+    page = recording_page
+
+    # Start on dashboard
+    page.goto(server_url)
+    wait_for_sidebar(page)
+    pause(page, 800)
+
+    # Navigate to immich via command palette (has multiple containers)
+    open_command_palette(page)
+    pause(page, 400)
+    slow_type(page, "#cmd-input", "immich", delay=100)
+    pause(page, 600)
+    page.keyboard.press("Enter")
+    page.wait_for_url("**/stack/immich", timeout=5000)
+    pause(page, 1500)
+
+    # Wait for containers list to load (so shell commands are available)
+    page.wait_for_selector("#containers-list button", timeout=10000)
+    pause(page, 800)
+
+    # Use command palette with fuzzy matching: "sh mach" -> "Shell: immich-machine-learning"
+    open_command_palette(page)
+    pause(page, 400)
+    slow_type(page, "#cmd-input", "sh mach", delay=100)
+    pause(page, 600)
+    page.keyboard.press("Enter")
+    pause(page, 1000)
+
+    # Wait for exec terminal to appear
+    page.wait_for_selector("#exec-terminal .xterm", timeout=10000)
+
+    # Smoothly scroll down to make the terminal visible
+    page.evaluate("""
+        const terminal = document.getElementById('exec-terminal');
+        if (terminal) {
+            terminal.scrollIntoView({ behavior: 'smooth', block: 'center' });
+        }
+    """)
+    pause(page, 1200)
+
+    # Run python version command
+    slow_type(page, "#exec-terminal .xterm-helper-textarea", "python3 --version", delay=60)
+    pause(page, 300)
+    page.keyboard.press("Enter")
+    pause(page, 1500)
+
+    # Blur the terminal to release focus (won't scroll)
+    page.evaluate("document.activeElement?.blur()")
+    pause(page, 500)
+
+    # Use command palette to switch to server container: "sh serv" -> "Shell: immich-server"
+    open_command_palette(page)
+    pause(page, 400)
+    slow_type(page, "#cmd-input", "sh serv", delay=100)
+    pause(page, 600)
+    page.keyboard.press("Enter")
+    pause(page, 1000)
+
+    # Wait for new terminal
+    page.wait_for_selector("#exec-terminal .xterm", timeout=10000)
+
+    # Scroll to terminal
+    page.evaluate("""
+        const terminal = document.getElementById('exec-terminal');
+        if (terminal) {
+            terminal.scrollIntoView({ behavior: 'smooth', block: 'center' });
+        }
+    """)
+    pause(page, 1200)
+
+    # Run ls command
+    slow_type(page, "#exec-terminal .xterm-helper-textarea", "ls /usr/src/app", delay=60)
+    pause(page, 300)
+    page.keyboard.press("Enter")
+    pause(page, 2000)

docs/demos/web/demo_stack.py

@@ -0,0 +1,101 @@
+"""Demo: Stack actions.
+
+Records a ~30 second demo showing:
+- Navigating to a stack page
+- Viewing compose file in Monaco editor
+- Triggering Restart action via command palette
+- Watching terminal output stream
+- Triggering Logs action
+
+Run: pytest docs/demos/web/demo_stack.py -v --no-cov
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import pytest
+from conftest import (
+    open_command_palette,
+    pause,
+    slow_type,
+    wait_for_sidebar,
+)
+
+if TYPE_CHECKING:
+    from playwright.sync_api import Page
+
+
+@pytest.mark.browser  # type: ignore[misc]
+def test_demo_stack(recording_page: Page, server_url: str) -> None:
+    """Record stack actions demo."""
+    page = recording_page
+
+    # Start on dashboard
+    page.goto(server_url)
+    wait_for_sidebar(page)
+    pause(page, 800)
+
+    # Navigate to grocy via command palette
+    open_command_palette(page)
+    pause(page, 400)
+    slow_type(page, "#cmd-input", "grocy", delay=100)
+    pause(page, 500)
+    page.keyboard.press("Enter")
+    page.wait_for_url("**/stack/grocy", timeout=5000)
+    pause(page, 1000)  # Show stack page
+
+    # Click on Compose File collapse to show the Monaco editor
+    # The collapse uses a checkbox input, click it via the parent collapse div
+    compose_collapse = page.locator(".collapse", has_text="Compose File").first
+    compose_collapse.locator("input[type=checkbox]").click(force=True)
+    pause(page, 500)
+
+    # Wait for Monaco editor to load and show content
+    page.wait_for_selector("#compose-editor .monaco-editor", timeout=10000)
+    pause(page, 2000)  # Let viewer see the compose file
+
+    # Smoothly scroll down to show more of the editor
+    page.evaluate("""
+        const editor = document.getElementById('compose-editor');
+        if (editor) {
+            editor.scrollIntoView({ behavior: 'smooth', block: 'center' });
+        }
+    """)
+    pause(page, 1200)  # Wait for smooth scroll animation
+
+    # Close the compose file section
+    compose_collapse.locator("input[type=checkbox]").click(force=True)
+    pause(page, 500)
+
+    # Open command palette for stack actions
+    open_command_palette(page)
+    pause(page, 400)
+
+    # Filter to Restart action
+    slow_type(page, "#cmd-input", "restart", delay=120)
+    pause(page, 600)
+
+    # Execute Restart
+    page.keyboard.press("Enter")
+    pause(page, 300)
+
+    # Wait for terminal to expand and show output
+    page.wait_for_selector("#terminal-output .xterm", timeout=5000)
+    pause(page, 2500)  # Let viewer see terminal streaming
+
+    # Open palette again for Logs
+    open_command_palette(page)
+    pause(page, 400)
+
+    # Filter to Logs action
+    slow_type(page, "#cmd-input", "logs", delay=120)
+    pause(page, 600)
+
+    # Execute Logs
+    page.keyboard.press("Enter")
+    pause(page, 300)
+
+    # Show log output
+    page.wait_for_selector("#terminal-output .xterm", timeout=5000)
+    pause(page, 2500)  # Final view of logs

docs/demos/web/demo_themes.py

@@ -0,0 +1,81 @@
+"""Demo: Theme switching.
+
+Records a ~15 second demo showing:
+- Opening theme picker via theme button
+- Live theme preview on arrow navigation
+- Selecting different themes
+- Theme persistence
+
+Run: pytest docs/demos/web/demo_themes.py -v --no-cov
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import pytest
+from conftest import (
+    pause,
+    slow_type,
+    wait_for_sidebar,
+)
+
+if TYPE_CHECKING:
+    from playwright.sync_api import Page
+
+
+@pytest.mark.browser  # type: ignore[misc]
+def test_demo_themes(recording_page: Page, server_url: str) -> None:
+    """Record theme switching demo."""
+    page = recording_page
+
+    # Start on dashboard
+    page.goto(server_url)
+    wait_for_sidebar(page)
+    pause(page, 1000)  # Show initial theme
+
+    # Click theme button to open theme picker
+    page.locator("#theme-btn").click()
+    page.wait_for_selector("#cmd-palette[open]", timeout=2000)
+    pause(page, 600)
+
+    # Arrow through many themes to show live preview effect
+    for _ in range(12):
+        page.keyboard.press("ArrowDown")
+        pause(page, 350)  # Show each preview
+
+    # Go back up through a few (land on valentine, not cyberpunk)
+    for _ in range(4):
+        page.keyboard.press("ArrowUp")
+        pause(page, 350)
+
+    # Select current theme with Enter
+    page.keyboard.press("Enter")
+    pause(page, 1000)
+
+    # Close palette with Escape
+    page.keyboard.press("Escape")
+    pause(page, 800)
+
+    # Open again and use search to find specific theme
+    page.locator("#theme-btn").click()
+    page.wait_for_selector("#cmd-palette[open]", timeout=2000)
+    pause(page, 400)
+
+    # Type to filter to a light theme (theme button pre-populates "theme:")
+    slow_type(page, "#cmd-input", "cup", delay=100)
+    pause(page, 500)
+    page.keyboard.press("Enter")
+    pause(page, 1000)
+
+    # Close and return to dark
+    page.keyboard.press("Escape")
+    pause(page, 500)
+    page.locator("#theme-btn").click()
+    page.wait_for_selector("#cmd-palette[open]", timeout=2000)
+    pause(page, 300)
+
+    slow_type(page, "#cmd-input", "dark", delay=100)
+    pause(page, 400)
+    page.keyboard.press("Enter")
+    pause(page, 800)

docs/demos/web/demo_workflow.py

@@ -0,0 +1,189 @@
+"""Demo: Full workflow.
+
+Records a comprehensive demo (~60 seconds) combining all major features:
+1. Console page: terminal with fastfetch, cf pull command
+2. Editor showing Compose Farm YAML config
+3. Command palette navigation to grocy stack
+4. Stack actions: up, logs
+5. Switch to dozzle stack via command palette, run update
+6. Dashboard overview
+7. Theme cycling via command palette
+
+This demo is used on the homepage and Web UI page as the main showcase.
+
+Run: pytest docs/demos/web/demo_workflow.py -v --no-cov
+"""
+
+from __future__ import annotations
+
+from typing import TYPE_CHECKING
+
+import pytest
+from conftest import open_command_palette, pause, slow_type, wait_for_sidebar
+
+if TYPE_CHECKING:
+    from playwright.sync_api import Page
+
+
+def _demo_console_terminal(page: Page, server_url: str) -> None:
+    """Demo part 1: Console page with terminal and editor."""
+    # Start on dashboard briefly
+    page.goto(server_url)
+    wait_for_sidebar(page)
+    pause(page, 800)
+
+    # Navigate to Console page via command palette
+    open_command_palette(page)
+    pause(page, 300)
+    slow_type(page, "#cmd-input", "cons", delay=100)
+    pause(page, 400)
+    page.keyboard.press("Enter")
+    page.wait_for_url("**/console", timeout=5000)
+    pause(page, 800)
+
+    # Wait for terminal to be ready
+    page.wait_for_selector("#console-terminal .xterm", timeout=10000)
+    pause(page, 1000)
+
+    # Run fastfetch first
+    slow_type(page, "#console-terminal .xterm-helper-textarea", "fastfetch", delay=60)
+    pause(page, 200)
+    page.keyboard.press("Enter")
+    pause(page, 2000)  # Wait for output
+
+    # Run cf pull on a stack to show Compose Farm in action
+    slow_type(page, "#console-terminal .xterm-helper-textarea", "cf pull grocy", delay=60)
+    pause(page, 200)
+    page.keyboard.press("Enter")
+    pause(page, 3000)  # Wait for pull output
+
+
+def _demo_config_editor(page: Page) -> None:
+    """Demo part 2: Show the Compose Farm config in editor."""
+    # Smoothly scroll down to show the Editor section
+    # Use JavaScript for smooth scrolling animation
+    page.evaluate("""
+        const editor = document.getElementById('console-editor');
+        if (editor) {
+            editor.scrollIntoView({ behavior: 'smooth', block: 'center' });
+        }
+    """)
+    pause(page, 1200)  # Wait for smooth scroll animation
+
+    # Wait for Monaco editor to load with config content
+    page.wait_for_selector("#console-editor .monaco-editor", timeout=10000)
+    pause(page, 2000)  # Let viewer see the Compose Farm config file
+
+
+def _demo_stack_actions(page: Page) -> None:
+    """Demo part 3: Navigate to stack and run actions."""
+    # Click on sidebar to take focus away from terminal, then use command palette
+    page.locator("#sidebar-stacks").click()
+    pause(page, 300)
+
+    # Navigate to grocy via command palette
+    open_command_palette(page)
+    pause(page, 300)
+    slow_type(page, "#cmd-input", "grocy", delay=100)
+    pause(page, 400)
+    page.keyboard.press("Enter")
+    page.wait_for_url("**/stack/grocy", timeout=5000)
+    pause(page, 1000)
+
+    # Open Compose File editor to show the compose.yaml
+    compose_collapse = page.locator(".collapse", has_text="Compose File").first
+    compose_collapse.locator("input[type=checkbox]").click(force=True)
+    pause(page, 500)
+
+    # Wait for Monaco editor to load and show content
+    page.wait_for_selector("#compose-editor .monaco-editor", timeout=10000)
+    pause(page, 2000)  # Let viewer see the compose file
+
+    # Close the compose file section
+    compose_collapse.locator("input[type=checkbox]").click(force=True)
+    pause(page, 500)
+
+    # Run Up action via command palette
+    open_command_palette(page)
+    pause(page, 300)
+    slow_type(page, "#cmd-input", "up", delay=100)
+    pause(page, 400)
+    page.keyboard.press("Enter")
+    pause(page, 200)
+
+    # Wait for terminal output
+    page.wait_for_selector("#terminal-output .xterm", timeout=5000)
+    pause(page, 2500)
+
+    # Show logs
+    open_command_palette(page)
+    pause(page, 300)
+    slow_type(page, "#cmd-input", "logs", delay=100)
+    pause(page, 400)
+    page.keyboard.press("Enter")
+    pause(page, 200)
+
+    page.wait_for_selector("#terminal-output .xterm", timeout=5000)
+    pause(page, 2500)
+
+    # Switch to dozzle via command palette (on nas for lower latency)
+    open_command_palette(page)
+    pause(page, 300)
+    slow_type(page, "#cmd-input", "dozzle", delay=100)
+    pause(page, 400)
+    page.keyboard.press("Enter")
+    page.wait_for_url("**/stack/dozzle", timeout=5000)
+    pause(page, 1000)
+
+    # Run update action
+    open_command_palette(page)
+    pause(page, 300)
+    slow_type(page, "#cmd-input", "upda", delay=100)
+    pause(page, 400)
+    page.keyboard.press("Enter")
+    pause(page, 200)
+
+    page.wait_for_selector("#terminal-output .xterm", timeout=5000)
+    pause(page, 2500)
+
+
+def _demo_dashboard_and_themes(page: Page, server_url: str) -> None:
+    """Demo part 4: Dashboard and theme cycling."""
+    # Navigate to dashboard via command palette
+    open_command_palette(page)
+    pause(page, 300)
+    slow_type(page, "#cmd-input", "dash", delay=100)
+    pause(page, 400)
+    page.keyboard.press("Enter")
+    page.wait_for_url(server_url, timeout=5000)
+    pause(page, 800)
+
+    # Scroll to top of page to ensure dashboard is fully visible
+    page.evaluate("window.scrollTo(0, 0)")
+    pause(page, 600)
+
+    # Open theme picker and arrow down to Dracula (shows live preview)
+    page.locator("#theme-btn").click()
+    page.wait_for_selector("#cmd-palette[open]", timeout=2000)
+    pause(page, 400)
+
+    # Arrow down through themes with live preview until we reach Dracula
+    for _ in range(19):
+        page.keyboard.press("ArrowDown")
+        pause(page, 180)
+
+    # Select Dracula theme and end on it
+    pause(page, 400)
+    page.keyboard.press("Enter")
+    pause(page, 1500)
+
+
+@pytest.mark.browser  # type: ignore[misc]
+def test_demo_workflow(recording_page: Page, server_url: str) -> None:
+    """Record full workflow demo."""
+    page = recording_page
+
+    _demo_console_terminal(page, server_url)
+    _demo_config_editor(page)
+    _demo_stack_actions(page)
+    _demo_dashboard_and_themes(page, server_url)

docs/demos/web/record.py

@@ -0,0 +1,259 @@
+#!/usr/bin/env python3
+"""Record all web UI demos.
+
+This script orchestrates recording of web UI demos using Playwright,
+then converts the WebM recordings to GIF format.
+
+Usage:
+    python docs/demos/web/record.py           # Record all demos
+    python docs/demos/web/record.py navigation  # Record specific demo
+
+Requirements:
+    - Playwright with Chromium: playwright install chromium
+    - ffmpeg for GIF conversion: apt install ffmpeg / brew install ffmpeg
+"""
+
+from __future__ import annotations
+
+import os
+import re
+import shutil
+import subprocess
+import sys
+from pathlib import Path
+
+from rich.console import Console
+
+console = Console()
+
+SCRIPT_DIR = Path(__file__).parent
+REPO_DIR = SCRIPT_DIR.parent.parent.parent
+OUTPUT_DIR = REPO_DIR / "docs" / "assets"
+
+DEMOS = [
+    "navigation",
+    "stack",
+    "themes",
+    "workflow",
+    "console",
+    "shell",
+    "live_stats",
+]
+
+# High-quality ffmpeg settings for VP8 encoding
+# See: https://github.com/microsoft/playwright/issues/10855
+# See: https://github.com/microsoft/playwright/issues/31424
+#
+# MAX_QUALITY: Lossless-like, largest files
+# BALANCED_QUALITY: ~43% file size, nearly indistinguishable quality
+MAX_QUALITY_ARGS = "-c:v vp8 -qmin 0 -qmax 0 -crf 0 -deadline best -speed 0 -b:v 0 -threads 0"
+BALANCED_QUALITY_ARGS = "-c:v vp8 -qmin 0 -qmax 10 -crf 4 -deadline best -speed 0 -b:v 0 -threads 0"
+
+# Choose which quality to use
+VIDEO_QUALITY_ARGS = MAX_QUALITY_ARGS
+
+
+def patch_playwright_video_quality() -> None:
+    """Patch Playwright's videoRecorder.js to use high-quality encoding settings."""
+    from playwright._impl._driver import compute_driver_executable  # noqa: PLC0415
+
+    # compute_driver_executable returns (node_path, cli_path)
+    result = compute_driver_executable()
+    node_path = result[0] if isinstance(result, tuple) else result
+    driver_path = Path(node_path).parent
+
+    video_recorder = driver_path / "package" / "lib" / "server" / "chromium" / "videoRecorder.js"
+
+    if not video_recorder.exists():
+        msg = f"videoRecorder.js not found at {video_recorder}"
+        raise FileNotFoundError(msg)
+
+    content = video_recorder.read_text()
+
+    # Check if already patched
+    if "deadline best" in content:
+        return  # Already patched
+
+    # Pattern to match the ffmpeg args line
+    pattern = (
+        r"-c:v vp8 -qmin \d+ -qmax \d+ -crf \d+ -deadline \w+ -speed \d+ -b:v \w+ -threads \d+"
+    )
+
+    if not re.search(pattern, content):
+        msg = "Could not find ffmpeg args pattern in videoRecorder.js"
+        raise ValueError(msg)
+
+    # Replace with high-quality settings
+    new_content = re.sub(pattern, VIDEO_QUALITY_ARGS, content)
+    video_recorder.write_text(new_content)
+    console.print("[green]Patched Playwright for high-quality video recording[/green]")
+
+
+def record_demo(name: str, index: int, total: int) -> Path | None:
+    """Run a single demo and return the video path."""
+    console.print(f"[cyan][{index}/{total}][/cyan] [green]Recording:[/green] web-{name}")
+
+    demo_file = SCRIPT_DIR / f"demo_{name}.py"
+    if not demo_file.exists():
+        console.print(f"[red]  Demo file not found: {demo_file}[/red]")
+        return None
+
+    # Create temp output dir for this recording
+    temp_dir = SCRIPT_DIR / ".recordings"
+    temp_dir.mkdir(exist_ok=True)
+
+    # Run pytest with video recording
+    # Set PYTHONPATH so conftest.py imports work
+    env = {**os.environ, "PYTHONPATH": str(SCRIPT_DIR)}
+    result = subprocess.run(
+        [
+            sys.executable,
+            "-m",
+            "pytest",
+            str(demo_file),
+            "-v",
+            "--no-cov",
+            "-x",  # Stop on first failure
+            f"--basetemp={temp_dir}",
+        ],
+        check=False,
+        cwd=REPO_DIR,
+        capture_output=True,
+        text=True,
+        env=env,
+    )
+
+    if result.returncode != 0:
+        console.print(f"[red]  Failed to record {name}[/red]")
+        console.print(result.stdout)
+        console.print(result.stderr)
+        return None
+
+    # Find the recorded video
+    videos = list(temp_dir.rglob("*.webm"))
+    if not videos:
+        console.print(f"[red]  No video found for {name}[/red]")
+        return None
+
+    # Use the most recent video
+    video = max(videos, key=lambda p: p.stat().st_mtime)
+    console.print(f"[green]  Recorded: {video.name}[/green]")
+    return video
+
+
+def convert_to_gif(webm_path: Path, output_name: str) -> Path:
+    """Convert WebM to GIF using ffmpeg with palette optimization."""
+    gif_path = OUTPUT_DIR / f"{output_name}.gif"
+    palette_path = webm_path.parent / "palette.png"
+
+    # Two-pass approach for better quality
+    # Pass 1: Generate palette
+    subprocess.run(
+        [  # noqa: S607
+            "ffmpeg",
+            "-y",
+            "-i",
+            str(webm_path),
+            "-vf",
+            "fps=10,scale=1280:-1:flags=lanczos,palettegen=stats_mode=diff",
+            str(palette_path),
+        ],
+        check=True,
+        capture_output=True,
+    )
+
+    # Pass 2: Generate GIF with palette
+    subprocess.run(
+        [  # noqa: S607
+            "ffmpeg",
+            "-y",
+            "-i",
+            str(webm_path),
+            "-i",
+            str(palette_path),
+            "-lavfi",
+            "fps=10,scale=1280:-1:flags=lanczos[x];[x][1:v]paletteuse=dither=bayer:bayer_scale=5:diff_mode=rectangle",
+            str(gif_path),
+        ],
+        check=True,
+        capture_output=True,
+    )
+
+    palette_path.unlink(missing_ok=True)
+    return gif_path
+
+
+def move_recording(video_path: Path, name: str) -> tuple[Path, Path]:
+    """Move WebM and convert to GIF, returning both paths."""
+    OUTPUT_DIR.mkdir(parents=True, exist_ok=True)
+
+    output_name = f"web-{name}"
+    webm_dest = OUTPUT_DIR / f"{output_name}.webm"
+
+    shutil.copy2(video_path, webm_dest)
+    console.print(f"[blue]  WebM: {webm_dest.relative_to(REPO_DIR)}[/blue]")
+
+    gif_path = convert_to_gif(video_path, output_name)
+    console.print(f"[blue]  GIF:  {gif_path.relative_to(REPO_DIR)}[/blue]")
+
+    return webm_dest, gif_path
+
+
+def cleanup() -> None:
+    """Clean up temporary recording files."""
+    temp_dir = SCRIPT_DIR / ".recordings"
+    if temp_dir.exists():
+        shutil.rmtree(temp_dir)
+
+
+def main() -> int:
+    """Record all web UI demos."""
+    console.print("[blue]Recording web UI demos...[/blue]")
+    console.print(f"Output directory: {OUTPUT_DIR}")
+    console.print()
+
+    # Patch Playwright for high-quality video recording
+    patch_playwright_video_quality()
+
+    # Determine which demos to record
+    if len(sys.argv) > 1:
+        demos_to_record = [d for d in sys.argv[1:] if d in DEMOS]
+        if not demos_to_record:
+            console.print(f"[red]Unknown demo(s). Available: {', '.join(DEMOS)}[/red]")
+            return 1
+    else:
+        demos_to_record = DEMOS
+
+    results: dict[str, tuple[Path | None, Path | None]] = {}
+
+    try:
+        for i, demo in enumerate(demos_to_record, 1):
+            video_path = record_demo(demo, i, len(demos_to_record))
+            if video_path:
+                webm, gif = move_recording(video_path, demo)
+                results[demo] = (webm, gif)
+            else:
+                results[demo] = (None, None)
+            console.print()
+    finally:
+        cleanup()
+
+    # Summary
+    console.print("[blue]=== Summary ===[/blue]")
+    success_count = sum(1 for w, _ in results.values() if w is not None)
+    console.print(f"Recorded: {success_count}/{len(demos_to_record)} demos")
+    console.print()
+
+    for demo, (webm, gif) in results.items():  # type: ignore[assignment]
+        status = "[green]OK[/green]" if webm else "[red]FAILED[/red]"
+        console.print(f"  {demo}: {status}")
+        if webm:
+            console.print(f"    {webm.relative_to(REPO_DIR)}")
+        if gif:
+            console.print(f"    {gif.relative_to(REPO_DIR)}")
+
+    return 0 if success_count == len(demos_to_record) else 1
+
+
+if __name__ == "__main__":
+    sys.exit(main())

docs/dev/docker-swarm-network.md

@@ -1,90 +0,0 @@
-# Docker Swarm Overlay Networks with Compose Farm
-
-Notes from testing Docker Swarm's attachable overlay networks as a way to get cross-host container networking while still using `docker compose`.
-
-## The Idea
-
-Docker Swarm overlay networks can be made "attachable", allowing regular `docker compose` containers (not just swarm services) to join them. This would give us:
-
-- Cross-host Docker DNS (containers find each other by name)
-- No need to publish ports for inter-container communication
-- Keep using `docker compose up` instead of `docker stack deploy`
-
-## Setup Steps
-
-```bash
-# On manager node
-docker swarm init --advertise-addr <manager-ip>
-
-# On worker nodes (use token from init output)
-docker swarm join --token <token> <manager-ip>:2377
-
-# Create attachable overlay network (on manager)
-docker network create --driver overlay --attachable my-network
-
-# In compose files, add the network
-networks:
-  my-network:
-    external: true
-```
-
-## Required Ports
-
-Docker Swarm requires these ports open **bidirectionally** between all nodes:
-
-| Port | Protocol | Purpose |
-|------|----------|---------|
-| 2377 | TCP | Cluster management |
-| 7946 | TCP + UDP | Node communication |
-| 4789 | UDP | Overlay network traffic (VXLAN) |
-
-## Test Results (2024-12-13)
-
-- docker-debian (192.168.1.66) as manager
-- dev-lxc (192.168.1.167) as worker
-
-### What worked
-
-- Swarm init and join
-- Overlay network creation
-- Nodes showed as Ready
-
-### What failed
-
-- Container on dev-lxc couldn't attach to overlay network
-- Error: `attaching to network failed... context deadline exceeded`
-- Cause: Port 7946 blocked from docker-debian → dev-lxc
-
-### Root cause
-
-Firewall on dev-lxc wasn't configured to allow swarm ports. Opening these ports requires sudo access on each node.
-
-## Conclusion
-
-Docker Swarm overlay networks are **not plug-and-play**. Requirements:
-
-1. Swarm init/join on all nodes
-2. Firewall rules on all nodes (needs sudo/root)
-3. All nodes must have bidirectional connectivity on 3 ports
-
-For a simpler alternative, consider:
-
-- **Tailscale**: VPN mesh, containers use host's Tailscale IP
-- **Host networking + published ports**: What compose-farm does today
-- **Keep dependent services together**: Avoid cross-host networking entirely
-
-## Future Work
-
-If we decide to support overlay networks:
-
-1. Add a `compose-farm network create` command that:
-   - Initializes swarm if needed
-   - Creates attachable overlay network
-   - Documents required firewall rules
-
-2. Add network config to compose-farm.yaml:
-   ```yaml
-   overlay_network: compose-farm-net
-   ```
-
-3. Auto-inject network into compose files (or document manual setup)

docs/docker-deployment.md

@@ -0,0 +1,101 @@
+---
+icon: lucide/container
+---
+
+# Docker Deployment
+
+Run the Compose Farm web UI in Docker.
+
+## Quick Start
+
+**1. Get the compose file:**
+
+```bash
+curl -O https://raw.githubusercontent.com/basnijholt/compose-farm/main/docker-compose.yml
+```
+
+**2. Generate `.env` file:**
+
+```bash
+cf config init-env
+```
+
+This auto-detects settings from your `compose-farm.yaml`:
+- `DOMAIN` from existing traefik labels
+- `CF_COMPOSE_DIR` from config
+- `CF_UID/GID/HOME/USER` from current user
+
+Review the output and edit if needed.
+
+**3. Set up SSH keys:**
+
+```bash
+docker compose run --rm cf ssh setup
+```
+
+**4. Start the web UI:**
+
+```bash
+docker compose up -d web
+```
+
+Open `http://localhost:9000` (or `https://compose-farm.example.com` if using Traefik).
+
+---
+
+## Configuration
+
+The `cf config init-env` command auto-detects most settings. After running it, review the generated `.env` file and edit if needed:
+
+```bash
+$EDITOR .env
+```
+
+### What init-env detects
+
+| Variable | How it's detected |
+|----------|-------------------|
+| `DOMAIN` | Extracted from traefik labels in your stacks |
+| `CF_COMPOSE_DIR` | From `compose_dir` in your config |
+| `CF_UID/GID/HOME/USER` | From current user (for NFS compatibility) |
+
+If auto-detection fails for any value, edit the `.env` file manually.
+
+### Glances Monitoring
+
+To show host CPU/memory stats in the dashboard, deploy [Glances](https://nicolargo.github.io/glances/) on your hosts. When running the web UI container, Compose Farm infers the local host from `CF_WEB_STACK` and uses the Glances container name for that host.
+
+See [Host Resource Monitoring](https://github.com/basnijholt/compose-farm#host-resource-monitoring-glances) in the README.
+
+---
+
+## Troubleshooting
+
+### SSH "Permission denied" or "Host key verification failed"
+
+Regenerate keys:
+
+```bash
+docker compose run --rm cf ssh setup
+```
+
+### Files created as root
+
+Add the non-root variables above and restart.
+
+---
+
+## All Environment Variables
+
+For advanced users, here's the complete reference:
+
+| Variable | Description | Default |
+|----------|-------------|---------|
+| `DOMAIN` | Domain for Traefik labels | *(required)* |
+| `CF_COMPOSE_DIR` | Compose files directory | `/opt/stacks` |
+| `CF_UID` / `CF_GID` | User/group ID | `0` (root) |
+| `CF_HOME` | Home directory | `/root` |
+| `CF_USER` | Username for SSH | `root` |
+| `CF_WEB_STACK` | Web UI stack name (enables self-update, local host inference) | *(none)* |
+| `CF_SSH_DIR` | SSH keys directory | `~/.ssh/compose-farm` |
+| `CF_XDG_CONFIG` | Config/backup directory | `~/.config/compose-farm` |

docs/getting-started.md

@@ -0,0 +1,340 @@
+---
+icon: lucide/rocket
+---
+
+# Getting Started
+
+This guide walks you through installing Compose Farm and setting up your first multi-host deployment.
+
+## Prerequisites
+
+Before you begin, ensure you have:
+
+- **[uv](https://docs.astral.sh/uv/)** (recommended) or Python 3.11+
+- **SSH key-based authentication** to your Docker hosts
+- **Docker and Docker Compose** installed on all target hosts
+- **Shared storage** for compose files (NFS, Syncthing, etc.)
+
+## Installation
+
+<video autoplay loop muted playsinline>
+  <source src="/assets/install.webm" type="video/webm">
+</video>
+
+### One-liner (recommended)
+
+```bash
+curl -fsSL https://compose-farm.nijho.lt/install | sh
+```
+
+This installs [uv](https://docs.astral.sh/uv/) if needed, then installs compose-farm.
+
+### Using uv
+
+If you already have [uv](https://docs.astral.sh/uv/) installed:
+
+```bash
+uv tool install compose-farm
+```
+
+### Using pip
+
+If you already have Python 3.11+ installed:
+
+```bash
+pip install compose-farm
+```
+
+### Using Docker
+
+```bash
+docker run --rm \
+  -v $SSH_AUTH_SOCK:/ssh-agent -e SSH_AUTH_SOCK=/ssh-agent \
+  -v ./compose-farm.yaml:/root/.config/compose-farm/compose-farm.yaml:ro \
+  ghcr.io/basnijholt/compose-farm up --all
+```
+
+**Running as non-root user** (recommended for NFS mounts):
+
+By default, containers run as root. To preserve file ownership on mounted volumes, set these environment variables in your `.env` file:
+
+```bash
+# Add to .env file (one-time setup)
+echo "CF_UID=$(id -u)" >> .env
+echo "CF_GID=$(id -g)" >> .env
+echo "CF_HOME=$HOME" >> .env
+echo "CF_USER=$USER" >> .env
+```
+
+Or use [direnv](https://direnv.net/) to auto-set these variables when entering the directory:
+```bash
+cp .envrc.example .envrc && direnv allow
+```
+
+This ensures files like `compose-farm-state.yaml` and web UI edits are owned by your user instead of root. The `CF_USER` variable is required for SSH to work when running as a non-root user.
+
+### Verify Installation
+
+```bash
+cf --version
+cf --help
+```
+
+## SSH Setup
+
+Compose Farm uses SSH to run commands on remote hosts. You need passwordless SSH access.
+
+### Option 1: SSH Agent (default)
+
+If you already have SSH keys loaded in your agent:
+
+```bash
+# Verify keys are loaded
+ssh-add -l
+
+# Test connection
+ssh user@192.168.1.10 "docker --version"
+```
+
+### Option 2: Dedicated Key (recommended for Docker)
+
+For persistent access when running in Docker:
+
+```bash
+# Generate and distribute key to all hosts
+cf ssh setup
+
+# Check status
+cf ssh status
+```
+
+This creates `~/.ssh/compose-farm/id_ed25519` and copies the public key to each host.
+
+## Shared Storage Setup
+
+Compose files must be accessible at the **same path** on all hosts. Common approaches:
+
+### NFS Mount
+
+```bash
+# On each Docker host
+sudo mount nas:/volume1/compose /opt/compose
+
+# Or add to /etc/fstab
+nas:/volume1/compose /opt/compose nfs defaults 0 0
+```
+
+### Directory Structure
+
+```
+/opt/compose/           # compose_dir in config
+├── plex/
+│   └── docker-compose.yml
+├── grafana/
+│   └── docker-compose.yml
+├── nextcloud/
+│   └── docker-compose.yml
+└── jellyfin/
+    └── docker-compose.yml
+```
+
+## Configuration
+
+### Create Config File
+
+Create `compose-farm.yaml` in the directory where you'll run commands. For example, if your stacks are in `/opt/stacks`, place the config there too:
+
+```bash
+cd /opt/stacks
+cf config init
+```
+
+Alternatively, use `~/.config/compose-farm/compose-farm.yaml` for a global config. You can also symlink a working directory config to the global location:
+
+```bash
+# Create config in your stacks directory, symlink to ~/.config
+cf config symlink /opt/stacks/compose-farm.yaml
+```
+
+This way, `cf` commands work from anywhere while the config lives with your stacks.
+
+#### Single host example
+
+```yaml
+# Where compose files are located (one folder per stack)
+compose_dir: /opt/stacks
+
+hosts:
+  local: localhost
+
+stacks:
+  plex: local
+  grafana: local
+  nextcloud: local
+```
+
+#### Multi-host example
+```yaml
+# Where compose files are located (same path on all hosts)
+compose_dir: /opt/compose
+
+# Define your Docker hosts
+hosts:
+  nuc:
+    address: 192.168.1.10
+    user: docker           # SSH user
+  hp:
+    address: 192.168.1.11
+    # user defaults to current user
+
+# Map stacks to hosts
+stacks:
+  plex: nuc
+  grafana: nuc
+  nextcloud: hp
+```
+
+Each entry in `stacks:` maps to a folder under `compose_dir` that contains a compose file.
+
+For cross-host HTTP routing, add Traefik labels and configure `traefik_file` (see [Traefik Integration](traefik.md)).
+### Validate Configuration
+
+```bash
+cf check --local
+```
+
+This validates syntax without SSH connections. For full validation:
+
+```bash
+cf check
+```
+
+## First Commands
+
+### Check Status
+
+```bash
+cf ps
+```
+
+Shows all configured stacks and their status.
+
+### Start All Stacks
+
+```bash
+cf up --all
+```
+
+Starts all stacks on their assigned hosts.
+
+### Start Specific Stacks
+
+```bash
+cf up plex grafana
+```
+
+### Apply Configuration
+
+The most powerful command - reconciles reality with your config:
+
+```bash
+cf apply --dry-run   # Preview changes
+cf apply             # Execute changes
+```
+
+This will:
+1. Start stacks in config but not running
+2. Migrate stacks on wrong host
+3. Stop stacks removed from config
+
+## Docker Network Setup
+
+If your stacks use an external Docker network:
+
+```bash
+# Create network on all hosts
+cf init-network
+
+# Or specific hosts
+cf init-network nuc hp
+```
+
+Default network: `mynetwork` with subnet `172.20.0.0/16`
+
+## Example Workflow
+
+### 1. Add a New Stack
+
+Create the compose file:
+
+```bash
+# On any host (shared storage)
+mkdir -p /opt/compose/gitea
+cat > /opt/compose/gitea/docker-compose.yml << 'EOF'
+services:
+  gitea:
+    image: docker.gitea.com/gitea:latest
+    container_name: gitea
+    environment:
+      - USER_UID=1000
+      - USER_GID=1000
+    volumes:
+      - /opt/config/gitea:/data
+      - /etc/timezone:/etc/timezone:ro
+      - /etc/localtime:/etc/localtime:ro
+    ports:
+      - "3000:3000"
+      - "2222:22"
+    restart: unless-stopped
+EOF
+```
+
+Add to config:
+
+```yaml
+stacks:
+  # ... existing stacks
+  gitea: nuc
+```
+
+Start the stack:
+
+```bash
+cf up gitea
+```
+
+### 2. Move a Stack to Another Host
+
+Edit `compose-farm.yaml`:
+
+```yaml
+stacks:
+  plex: hp  # Changed from nuc
+```
+
+Apply the change:
+
+```bash
+cf up plex
+# Automatically: down on nuc, up on hp
+```
+
+Or use apply to reconcile everything:
+
+```bash
+cf apply
+```
+
+### 3. Update All Stacks
+
+```bash
+cf update --all
+# Only recreates containers if images changed
+```
+
+## Next Steps
+
+- [Configuration Reference](configuration.md) - All config options
+- [Commands Reference](commands.md) - Full CLI documentation
+- [Traefik Integration](traefik.md) - Multi-host routing
+- [Best Practices](best-practices.md) - Tips and limitations

docs/index.md

@@ -0,0 +1,167 @@
+---
+icon: lucide/server
+---
+
+# Compose Farm
+
+A minimal CLI tool to run Docker Compose commands across multiple hosts via SSH.
+
+## What is Compose Farm?
+
+Compose Farm lets you manage Docker Compose stacks across multiple machines from a single command line. Think [Dockge](https://dockge.kuma.pet/) but with a CLI and web interface, designed for multi-host deployments.
+
+Define which stacks run where in one YAML file, then use `cf apply` to make reality match your configuration.
+It also works great on a single host with one folder per stack; just map stacks to `localhost`.
+
+## Quick Demo
+
+**CLI:**
+<video autoplay loop muted playsinline>
+  <source src="/assets/quickstart.webm" type="video/webm">
+</video>
+
+**[Web UI](web-ui.md):**
+<video autoplay loop muted playsinline>
+  <source src="/assets/web-workflow.webm" type="video/webm">
+</video>
+
+## Why Compose Farm?
+
+| Problem | Compose Farm Solution |
+|---------|----------------------|
+| 100+ containers on one machine | Distribute across multiple hosts |
+| Kubernetes too complex | Just SSH + docker compose |
+| Swarm in maintenance mode | Zero infrastructure changes |
+| Manual SSH for each host | Single command for all |
+
+**It's a convenience wrapper, not a new paradigm.** Your existing `docker-compose.yml` files work unchanged.
+
+## Quick Start
+
+### Single host
+
+No SSH, shared storage, or Traefik file-provider required.
+
+```yaml
+# compose-farm.yaml
+compose_dir: /opt/stacks
+
+hosts:
+  local: localhost
+
+stacks:
+  plex: local
+  jellyfin: local
+  traefik: local
+```
+
+```bash
+cf apply  # Start/stop stacks to match config
+```
+
+### Multi-host
+
+Requires SSH plus a shared `compose_dir` path on all hosts (NFS or sync).
+
+```yaml
+# compose-farm.yaml
+compose_dir: /opt/compose
+
+hosts:
+  server-1:
+    address: 192.168.1.10
+  server-2:
+    address: 192.168.1.11
+
+stacks:
+  plex: server-1
+  jellyfin: server-2
+  grafana: server-1
+```
+
+```bash
+cf apply  # Stacks start, migrate, or stop as needed
+```
+
+Each entry in `stacks:` maps to a folder under `compose_dir` that contains a compose file.
+
+For cross-host HTTP routing, add Traefik labels and configure `traefik_file` to generate file-provider config.
+### Installation
+
+```bash
+uv tool install compose-farm
+# or
+pip install compose-farm
+```
+
+### Configuration
+
+Create `compose-farm.yaml` in the directory where you'll run commands (e.g., `/opt/stacks`), or in `~/.config/compose-farm/`:
+
+```yaml
+compose_dir: /opt/compose
+
+hosts:
+  nuc:
+    address: 192.168.1.10
+    user: docker
+  hp:
+    address: 192.168.1.11
+
+stacks:
+  plex: nuc
+  grafana: nuc
+  nextcloud: hp
+```
+
+See [Configuration](configuration.md) for all options and the full search order.
+
+### Usage
+
+```bash
+# Make reality match config
+cf apply
+
+# Start specific stacks
+cf up plex grafana
+
+# Check status
+cf ps
+
+# View logs
+cf logs -f plex
+```
+
+## Key Features
+
+- **Declarative configuration**: One YAML defines where everything runs
+- **Auto-migration**: Change a host assignment, run `cf up`, stack moves automatically
+
+<video autoplay loop muted playsinline>
+  <source src="/assets/migration.webm" type="video/webm">
+</video>
+- **Parallel execution**: Multiple stacks start/stop concurrently
+- **State tracking**: Knows which stacks are running where
+- **Traefik integration**: Generate file-provider config for cross-host routing
+- **Zero changes**: Your compose files work as-is
+
+## Requirements
+
+- [uv](https://docs.astral.sh/uv/) (recommended) or Python 3.11+
+- SSH key-based authentication to your Docker hosts
+- Docker and Docker Compose on all target hosts
+- Shared storage (compose files at same path on all hosts)
+
+## Documentation
+
+- [Getting Started](getting-started.md) - Installation and first steps
+- [Configuration](configuration.md) - All configuration options
+- [Commands](commands.md) - CLI reference
+- [Web UI](web-ui.md) - Browser-based management interface
+- [Architecture](architecture.md) - How it works under the hood
+- [Traefik Integration](traefik.md) - Multi-host routing setup
+- [Best Practices](best-practices.md) - Tips and limitations
+
+## License
+
+MIT

docs/install

@@ -0,0 +1,29 @@
+#!/bin/sh
+# Compose Farm bootstrap script
+# Usage: curl -fsSL https://compose-farm.nijho.lt/install | sh
+#
+# This script installs uv (if needed) and then installs compose-farm as a uv tool.
+
+set -e
+
+if ! command -v uv >/dev/null 2>&1; then
+    echo "uv is not installed. Installing..."
+    curl -LsSf https://astral.sh/uv/install.sh | sh
+    echo "uv installation complete!"
+    echo ""
+
+    if [ -x ~/.local/bin/uv ]; then
+        ~/.local/bin/uv tool install compose-farm
+    else
+        echo "Please restart your shell and run this script again"
+        echo ""
+        exit 0
+    fi
+else
+    uv tool install compose-farm
+fi
+
+echo ""
+echo "compose-farm is installed!"
+echo "Run 'cf --help' to get started."
+echo "If 'cf' is not found, restart your shell or run: source ~/.bashrc"

docs/javascripts/video-fix.js

@@ -0,0 +1,21 @@
+// Fix Safari video autoplay issues
+(function() {
+  function initVideos() {
+    document.querySelectorAll('video[autoplay]').forEach(function(video) {
+      video.load();
+      video.play().catch(function() {});
+    });
+  }
+
+  // For initial page load (needed for Chrome)
+  if (document.readyState === 'loading') {
+    document.addEventListener('DOMContentLoaded', initVideos);
+  } else {
+    initVideos();
+  }
+
+  // For MkDocs instant navigation (needed for Safari)
+  if (typeof document$ !== 'undefined') {
+    document$.subscribe(initVideos);
+  }
+})();

docs/overrides/partials/integrations/analytics/custom.html

@@ -0,0 +1,6 @@
+<!-- Privacy-friendly analytics by Plausible -->
+<script async src="https://plausible.nijho.lt/js/pa-NRX7MolONWKTUREJpAjkB.js"></script>
+<script>
+  window.plausible=window.plausible||function(){(plausible.q=plausible.q||[]).push(arguments)},plausible.init=plausible.init||function(i){plausible.o=i||{}};
+  plausible.init()
+</script>

docs/reddit-post.md

@@ -0,0 +1,79 @@
+# Title options
+
+- Multi-host Docker Compose without Kubernetes or file changes
+- I built a CLI to run Docker Compose across hosts. Zero changes to your files.
+- I made a CLI to run Docker Compose across multiple hosts without Kubernetes or Swarm
+---
+
+I've been running 100+ Docker Compose stacks on a single machine, and it kept running out of memory. I needed to spread stacks across multiple hosts, but:
+
+- **Kubernetes** felt like overkill. I don't need pods, ingress controllers, or 10x more YAML.
+- **Docker Swarm** is basically in maintenance mode.
+- Both require rewriting my compose files.
+
+So I built **Compose Farm**, a simple CLI that runs `docker compose` commands over SSH. No agents, no cluster setup, no changes to your existing compose files.
+
+## How it works
+
+One YAML file maps stacks to hosts:
+
+```yaml
+compose_dir: /opt/stacks
+
+hosts:
+  nuc: 192.168.1.10
+  hp: 192.168.1.11
+
+stacks:
+  plex: nuc
+  jellyfin: hp
+  grafana: nuc
+  nextcloud: nuc
+```
+
+Then just:
+
+```bash
+cf up plex        # runs on nuc via SSH
+cf apply          # makes config state match desired state on all hosts (like Terraform apply)
+cf up --all       # starts everything on their assigned hosts
+cf logs -f plex   # streams logs
+cf ps             # shows status across all hosts
+```
+
+## Auto-migration
+
+Change a stack's host in the config and run `cf up`. It stops the stack on the old host and starts it on the new one. No manual SSH needed.
+
+```yaml
+# Before
+plex: nuc
+
+# After (just change this)
+plex: hp
+```
+
+```bash
+cf up plex  # migrates automatically
+```
+
+## Requirements
+
+- SSH key auth to your hosts
+- Same paths on all hosts (I use NFS from my NAS)
+- That's it. No agents, no daemons.
+
+## What it doesn't do
+
+- No high availability (if a host goes down, stacks don't auto-migrate)
+- No overlay networking (containers on different hosts can't talk via Docker DNS)
+- No health checks or automatic restarts
+
+It's a convenience wrapper around `docker compose` + SSH. If you need failover or cross-host container networking, you probably do need Swarm or Kubernetes.
+
+## Links
+
+- GitHub: https://github.com/basnijholt/compose-farm
+- Install: `uv tool install compose-farm` or `pip install compose-farm`
+
+Happy to answer questions or take feedback!

docs/traefik.md

@@ -0,0 +1,384 @@
+---
+icon: lucide/globe
+---
+
+# Traefik Integration
+
+Compose Farm can generate Traefik file-provider configuration for routing traffic across multiple hosts.
+
+## The Problem
+
+When you run Traefik on one host but stacks on others, Traefik's docker provider can't see remote containers. The file provider bridges this gap.
+
+```
+                    Internet
+                        │
+                        ▼
+┌─────────────────────────────────────────────────────────────┐
+│                     Host: nuc                                │
+│                                                             │
+│  ┌─────────┐                                                │
+│  │ Traefik │◄─── Docker provider sees local containers      │
+│  │         │                                                │
+│  │         │◄─── File provider sees remote stacks           │
+│  └────┬────┘     (from compose-farm.yml)                    │
+│       │                                                     │
+└───────┼─────────────────────────────────────────────────────┘
+        │
+        ├────────────────────┐
+        │                    │
+        ▼                    ▼
+┌───────────────┐    ┌───────────────┐
+│   Host: hp    │    │  Host: nas    │
+│               │    │               │
+│  plex:32400   │    │ jellyfin:8096 │
+└───────────────┘    └───────────────┘
+```
+
+## How It Works
+
+1. Your compose files have standard Traefik labels
+2. Compose Farm reads labels and generates file-provider config
+3. Traefik watches the generated file
+4. Traffic routes to remote stacks via host IP + published port
+
+## Setup
+
+### Step 1: Configure Traefik File Provider
+
+Add directory watching to your Traefik config:
+
+```yaml
+# traefik.yml or docker-compose.yml command
+providers:
+  file:
+    directory: /opt/traefik/dynamic.d
+    watch: true
+```
+
+Or via command line:
+
+```yaml
+services:
+  traefik:
+    command:
+      - --providers.file.directory=/dynamic.d
+      - --providers.file.watch=true
+    volumes:
+      - /opt/traefik/dynamic.d:/dynamic.d:ro
+```
+
+### Step 2: Add Traefik Labels to Services
+
+Your compose files use standard Traefik labels:
+
+```yaml
+# /opt/compose/plex/docker-compose.yml
+services:
+  plex:
+    image: lscr.io/linuxserver/plex
+    ports:
+      - "32400:32400"  # IMPORTANT: Must publish port!
+    labels:
+      - traefik.enable=true
+      - traefik.http.routers.plex.rule=Host(`plex.example.com`)
+      - traefik.http.routers.plex.entrypoints=websecure
+      - traefik.http.routers.plex.tls.certresolver=letsencrypt
+      - traefik.http.services.plex.loadbalancer.server.port=32400
+```
+
+**Important:** Services must publish ports for cross-host routing. Traefik connects via `host_ip:published_port`.
+
+### Step 3: Generate File Provider Config
+
+```bash
+cf traefik-file --all -o /opt/traefik/dynamic.d/compose-farm.yml
+```
+
+This generates:
+
+```yaml
+# /opt/traefik/dynamic.d/compose-farm.yml
+http:
+  routers:
+    plex:
+      rule: Host(`plex.example.com`)
+      entryPoints:
+        - websecure
+      tls:
+        certResolver: letsencrypt
+      service: plex
+  services:
+    plex:
+      loadBalancer:
+        servers:
+          - url: http://192.168.1.11:32400
+```
+
+## Auto-Regeneration
+
+Configure automatic regeneration in `compose-farm.yaml`:
+
+```yaml
+compose_dir: /opt/compose
+traefik_file: /opt/traefik/dynamic.d/compose-farm.yml
+traefik_stack: traefik
+
+hosts:
+  nuc:
+    address: 192.168.1.10
+  hp:
+    address: 192.168.1.11
+
+stacks:
+  traefik: nuc      # Traefik runs here
+  plex: hp          # Routed via file-provider
+  grafana: hp
+```
+
+With `traefik_file` set, these commands auto-regenerate the config:
+- `cf up`
+- `cf down`
+- `cf update`
+- `cf apply`
+
+### traefik_stack Option
+
+When set, stacks on the **same host as Traefik** are skipped in file-provider output. Traefik's docker provider handles them directly.
+
+```yaml
+traefik_stack: traefik  # traefik runs on nuc
+stacks:
+  traefik: nuc            # NOT in file-provider (docker provider)
+  portainer: nuc          # NOT in file-provider (docker provider)
+  plex: hp                # IN file-provider (cross-host)
+```
+
+## Label Syntax
+
+### Routers
+
+```yaml
+labels:
+  # Basic router
+  - traefik.http.routers.myapp.rule=Host(`app.example.com`)
+  - traefik.http.routers.myapp.entrypoints=websecure
+
+  # With TLS
+  - traefik.http.routers.myapp.tls=true
+  - traefik.http.routers.myapp.tls.certresolver=letsencrypt
+
+  # With middleware
+  - traefik.http.routers.myapp.middlewares=auth@file
+```
+
+### Services
+
+```yaml
+labels:
+  # Load balancer port
+  - traefik.http.services.myapp.loadbalancer.server.port=8080
+
+  # Health check
+  - traefik.http.services.myapp.loadbalancer.healthcheck.path=/health
+```
+
+### Middlewares
+
+Middlewares should be defined in a separate file (not generated by Compose Farm):
+
+```yaml
+# /opt/traefik/dynamic.d/middlewares.yml
+http:
+  middlewares:
+    auth:
+      basicAuth:
+        users:
+          - "user:$apr1$..."
+```
+
+Reference in labels:
+
+```yaml
+labels:
+  - traefik.http.routers.myapp.middlewares=auth@file
+```
+
+## Variable Substitution
+
+Labels can use environment variables:
+
+```yaml
+labels:
+  - traefik.http.routers.myapp.rule=Host(`${DOMAIN}`)
+```
+
+Compose Farm resolves variables from:
+1. Stack's `.env` file
+2. Current environment
+
+```bash
+# /opt/compose/myapp/.env
+DOMAIN=app.example.com
+```
+
+## Port Resolution
+
+Compose Farm determines the target URL from published ports:
+
+```yaml
+ports:
+  - "8080:80"           # Uses 8080
+  - "192.168.1.11:8080:80"  # Uses 8080 on specific IP
+```
+
+If no suitable port is found, a warning is shown.
+
+## Complete Example
+
+### compose-farm.yaml
+
+```yaml
+compose_dir: /opt/compose
+traefik_file: /opt/traefik/dynamic.d/compose-farm.yml
+traefik_stack: traefik
+
+hosts:
+  nuc:
+    address: 192.168.1.10
+  hp:
+    address: 192.168.1.11
+  nas:
+    address: 192.168.1.100
+
+stacks:
+  traefik: nuc
+  plex: hp
+  jellyfin: nas
+  grafana: nuc
+  nextcloud: nuc
+```
+
+### /opt/compose/plex/docker-compose.yml
+
+```yaml
+services:
+  plex:
+    image: lscr.io/linuxserver/plex
+    container_name: plex
+    ports:
+      - "32400:32400"
+    labels:
+      - traefik.enable=true
+      - traefik.http.routers.plex.rule=Host(`plex.example.com`)
+      - traefik.http.routers.plex.entrypoints=websecure
+      - traefik.http.routers.plex.tls.certresolver=letsencrypt
+      - traefik.http.services.plex.loadbalancer.server.port=32400
+    # ... other config
+```
+
+### Generated compose-farm.yml
+
+```yaml
+http:
+  routers:
+    plex:
+      rule: Host(`plex.example.com`)
+      entryPoints:
+        - websecure
+      tls:
+        certResolver: letsencrypt
+      service: plex
+    jellyfin:
+      rule: Host(`jellyfin.example.com`)
+      entryPoints:
+        - websecure
+      tls:
+        certResolver: letsencrypt
+      service: jellyfin
+
+  services:
+    plex:
+      loadBalancer:
+        servers:
+          - url: http://192.168.1.11:32400
+    jellyfin:
+      loadBalancer:
+        servers:
+          - url: http://192.168.1.100:8096
+```
+
+Note: `grafana` and `nextcloud` are NOT in the file because they're on the same host as Traefik (`nuc`).
+
+## Combining with Existing Config
+
+If you have existing Traefik dynamic config:
+
+```bash
+# Move existing config to directory
+mkdir -p /opt/traefik/dynamic.d
+mv /opt/traefik/dynamic.yml /opt/traefik/dynamic.d/manual.yml
+
+# Generate Compose Farm config
+cf traefik-file --all -o /opt/traefik/dynamic.d/compose-farm.yml
+
+# Update Traefik to watch directory
+# --providers.file.directory=/dynamic.d
+```
+
+Traefik merges all YAML files in the directory.
+
+## Troubleshooting
+
+### Stack Not Accessible
+
+1. **Check port is published:**
+   ```yaml
+   ports:
+     - "8080:80"  # Must be published, not just exposed
+   ```
+
+2. **Check label syntax:**
+   ```bash
+   cf check mystack
+   ```
+
+3. **Verify generated config:**
+   ```bash
+   cf traefik-file mystack
+   ```
+
+4. **Check Traefik logs:**
+   ```bash
+   docker logs traefik
+   ```
+
+### Config Not Regenerating
+
+1. **Verify traefik_file is set:**
+   ```bash
+   cf config show | grep traefik
+   ```
+
+2. **Check file permissions:**
+   ```bash
+   ls -la /opt/traefik/dynamic.d/
+   ```
+
+3. **Manually regenerate:**
+   ```bash
+   cf traefik-file --all -o /opt/traefik/dynamic.d/compose-farm.yml
+   ```
+
+### Variable Not Resolved
+
+1. **Check .env file exists:**
+   ```bash
+   cat /opt/compose/myservice/.env
+   ```
+
+2. **Test variable resolution:**
+   ```bash
+   cd /opt/compose/myservice
+   docker compose config
+   ```

docs/truenas-nested-nfs.md

@@ -0,0 +1,169 @@
+# TrueNAS NFS: Accessing Child ZFS Datasets
+
+When NFS-exporting a parent ZFS dataset on TrueNAS, child datasets appear as **empty directories** to NFS clients. This document explains the problem and provides a workaround.
+
+## The Problem
+
+TrueNAS structures storage as ZFS datasets. A common pattern is:
+
+```
+tank/data              <- parent dataset (NFS exported)
+tank/data/app1         <- child dataset
+tank/data/app2         <- child dataset
+```
+
+When you create an NFS share for `tank/data`, clients mount it and see the `app1/` and `app2/` directories—but they're empty. This happens because each ZFS dataset is a separate filesystem, and NFS doesn't traverse into child filesystems by default.
+
+## The Solution: `crossmnt`
+
+The NFS `crossmnt` export option tells the server to allow clients to traverse into child filesystems. However, TrueNAS doesn't expose this option in the UI.
+
+### Workaround Script
+
+This Python script injects `crossmnt` into `/etc/exports`:
+
+```python
+#!/usr/bin/env python3
+"""
+Add crossmnt to TrueNAS NFS exports for child dataset visibility.
+
+Usage: fix-nfs-crossmnt.py /mnt/pool/dataset
+
+Setup:
+  1. scp fix-nfs-crossmnt.py root@truenas.local:/root/
+  2. chmod +x /root/fix-nfs-crossmnt.py
+  3. Test: /root/fix-nfs-crossmnt.py /mnt/pool/dataset
+  4. Add cron job: TrueNAS UI > System > Advanced > Cron Jobs
+     Command: /root/fix-nfs-crossmnt.py /mnt/pool/dataset
+     Schedule: */5 * * * *
+"""
+
+import re
+import subprocess
+import sys
+from pathlib import Path
+
+EXPORTS_FILE = Path("/etc/exports")
+
+
+def main():
+    if len(sys.argv) != 2:
+        print(f"Usage: {sys.argv[0]} /mnt/pool/dataset", file=sys.stderr)
+        return 1
+
+    export_path = sys.argv[1]
+    content = EXPORTS_FILE.read_text()
+
+    if f'"{export_path}"' not in content:
+        print(f"ERROR: {export_path} not found in {EXPORTS_FILE}", file=sys.stderr)
+        return 1
+
+    lines = content.splitlines()
+    result = []
+    in_block = False
+    modified = False
+
+    for line in lines:
+        if f'"{export_path}"' in line:
+            in_block = True
+        elif line.startswith('"'):
+            in_block = False
+
+        if in_block and line[:1] in (" ", "\t") and "crossmnt" not in line:
+            line = re.sub(r"\)(\\\s*)?$", r",crossmnt)\1", line)
+            modified = True
+
+        result.append(line)
+
+    if not modified:
+        return 0  # Already applied
+
+    EXPORTS_FILE.write_text("\n".join(result) + "\n")
+    subprocess.run(["exportfs", "-ra"], check=True)
+    print(f"Added crossmnt to {export_path}")
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())
+```
+
+## Setup Instructions
+
+### 1. Copy the script to TrueNAS
+
+```bash
+scp fix-nfs-crossmnt.py root@truenas.local:/root/
+ssh root@truenas.local chmod +x /root/fix-nfs-crossmnt.py
+```
+
+### 2. Test manually
+
+```bash
+ssh root@truenas.local
+
+# Run the script
+/root/fix-nfs-crossmnt.py /mnt/tank/data
+
+# Verify crossmnt was added
+cat /etc/exports
+```
+
+You should see `,crossmnt` added to the client options:
+
+```
+"/mnt/tank/data"\
+    192.168.1.10(sec=sys,rw,no_subtree_check,crossmnt)\
+    192.168.1.11(sec=sys,rw,no_subtree_check,crossmnt)
+```
+
+### 3. Verify on NFS client
+
+```bash
+# Before: empty directory
+ls /mnt/data/app1/
+# (nothing)
+
+# After: actual contents visible
+ls /mnt/data/app1/
+# config.yaml  data/  logs/
+```
+
+### 4. Make it persistent
+
+TrueNAS regenerates `/etc/exports` when you modify NFS shares in the UI. To survive this, set up a cron job:
+
+1. Go to **TrueNAS UI → System → Advanced → Cron Jobs → Add**
+2. Configure:
+   - **Description:** Fix NFS crossmnt
+   - **Command:** `/root/fix-nfs-crossmnt.py /mnt/tank/data`
+   - **Run As User:** root
+   - **Schedule:** `*/5 * * * *` (every 5 minutes)
+   - **Enabled:** checked
+3. Save
+
+The script is idempotent—it only modifies the file if `crossmnt` is missing, and skips the write entirely if already applied.
+
+## How It Works
+
+1. Parses `/etc/exports` to find the specified export block
+2. Adds `,crossmnt` before the closing `)` on each client line
+3. Writes the file only if changes were made
+4. Runs `exportfs -ra` to reload the NFS configuration
+
+## Why Not Use SMB Instead?
+
+SMB handles child datasets seamlessly, but:
+
+- NFS is simpler for Linux-to-Linux with matching UIDs
+- SMB requires more complex permission mapping for Docker volumes
+- Many existing setups already use NFS
+
+## Related Links
+
+- [TrueNAS Forum: Add crossmnt option to NFS exports](https://forums.truenas.com/t/add-crossmnt-option-to-nfs-exports/10573)
+- [exports(5) man page](https://man7.org/linux/man-pages/man5/exports.5.html) - see `crossmnt` option
+
+## Tested On
+
+- TrueNAS SCALE 24.10

docs/truenas-nfs-root-squash.md

@@ -0,0 +1,65 @@
+# TrueNAS NFS: Disabling Root Squash
+
+When running Docker containers on NFS-mounted storage, containers that run as root will fail to write files unless root squash is disabled. This document explains the problem and solution.
+
+## The Problem
+
+By default, NFS uses "root squash" which maps the root user (UID 0) on clients to `nobody` on the server. This is a security feature to prevent remote root users from having root access to the NFS server's files.
+
+However, many Docker containers run as root internally. When these containers try to write to NFS-mounted volumes, the writes fail with "Permission denied" because the NFS server sees them as `nobody`, not `root`.
+
+Example error in container logs:
+```
+System.UnauthorizedAccessException: Access to the path '/data' is denied.
+Error: EACCES: permission denied, mkdir '/app/data'
+```
+
+## The Solution
+
+In TrueNAS, configure the NFS share to map remote root to local root:
+
+### TrueNAS SCALE UI
+
+1. Go to **Shares → NFS**
+2. Edit your share
+3. Under **Advanced Options**:
+   - **Maproot User**: `root`
+   - **Maproot Group**: `wheel`
+4. Save
+
+### Result in /etc/exports
+
+```
+"/mnt/pool/data"\
+    192.168.1.25(sec=sys,rw,no_root_squash,no_subtree_check)\
+    192.168.1.26(sec=sys,rw,no_root_squash,no_subtree_check)
+```
+
+The `no_root_squash` option means remote root is treated as root on the server.
+
+## Why `wheel`?
+
+On FreeBSD/TrueNAS, the root user's primary group is `wheel` (GID 0), not `root` like on Linux. So `root:wheel` = `0:0`.
+
+## Security Considerations
+
+Disabling root squash means any machine that can mount the NFS share has full root access to those files. This is acceptable when:
+
+- The NFS clients are on a trusted private network
+- Only known hosts (by IP) are allowed to mount the share
+- The data isn't security-critical
+
+For home lab setups with Docker containers, this is typically fine.
+
+## Alternative: Run Containers as Non-Root
+
+If you prefer to keep root squash enabled, you can run containers as a non-root user:
+
+1. **LinuxServer.io images**: Set `PUID=1000` and `PGID=1000` environment variables
+2. **Other images**: Add `user: "1000:1000"` to the compose service
+
+However, not all containers support running as non-root (they may need to bind to privileged ports, create system directories, etc.).
+
+## Tested On
+
+- TrueNAS SCALE 24.10

docs/web-ui.md

@@ -0,0 +1,154 @@
+---
+icon: lucide/layout-dashboard
+---
+
+# Web UI
+
+Compose Farm includes a web interface for managing stacks from your browser. Start it with:
+
+```bash
+cf web
+```
+
+Then open [http://localhost:8000](http://localhost:8000).
+
+## Features
+
+### Full Workflow
+
+Console terminal, config editor, stack navigation, actions (up, logs, update), dashboard overview, and theme switching - all in one flow.
+
+<video autoplay loop muted playsinline>
+  <source src="/assets/web-workflow.webm" type="video/webm">
+</video>
+
+### Stack Actions
+
+Navigate to any stack and use the command palette to trigger actions like restart, pull, update, or view logs. Output streams in real-time via WebSocket.
+
+<video autoplay loop muted playsinline>
+  <source src="/assets/web-stack.webm" type="video/webm">
+</video>
+
+### Theme Switching
+
+35 themes available via the command palette. Type `theme:` to filter, then use arrow keys to preview themes live before selecting.
+
+<video autoplay loop muted playsinline>
+  <source src="/assets/web-themes.webm" type="video/webm">
+</video>
+
+### Command Palette
+
+Press `Ctrl+K` (or `Cmd+K` on macOS) to open the command palette. Use fuzzy search to quickly navigate, trigger actions, or change themes.
+
+<video autoplay loop muted playsinline>
+  <source src="/assets/web-navigation.webm" type="video/webm">
+</video>
+
+## Pages
+
+### Dashboard (`/`)
+
+- Stack overview with status indicators
+- Host statistics (CPU, memory, disk, load via Glances)
+- Pending operations (migrations, orphaned stacks)
+- Quick actions via command palette
+
+### Live Stats (`/live-stats`)
+
+Real-time container monitoring across all hosts, powered by [Glances](https://nicolargo.github.io/glances/).
+
+- **Live metrics**: CPU, memory, network I/O for every container
+- **Auto-refresh**: Updates every 3 seconds (pauses when dropdown menus are open)
+- **Filtering**: Type to filter containers by name, stack, host, or image
+- **Sorting**: Click column headers to sort by any metric
+- **Update detection**: Shows when container images have updates available
+
+<video autoplay loop muted playsinline>
+  <source src="/assets/web-live_stats.webm" type="video/webm">
+</video>
+
+#### Requirements
+
+Live Stats requires Glances to be deployed on all hosts:
+
+1. Add `glances_stack: glances` to your `compose-farm.yaml`
+2. Deploy a Glances stack that runs on all hosts (see [example](https://github.com/basnijholt/compose-farm/tree/main/examples/glances))
+3. Glances must expose its REST API on port 61208
+
+### Stack Detail (`/stack/{name}`)
+
+- Compose file editor (Monaco)
+- Environment file editor
+- Action buttons: Up, Down, Restart, Update, Pull, Logs
+- Container shell access (exec into running containers)
+- Terminal output for running commands
+
+Files are automatically backed up before saving to `~/.config/compose-farm/backups/`.
+
+### Console (`/console`)
+
+- Full shell access to any host
+- File editor for remote files
+- Monaco editor with syntax highlighting
+
+<video autoplay loop muted playsinline>
+  <source src="/assets/web-console.webm" type="video/webm">
+</video>
+
+### Container Shell
+
+Click the Shell button on any running container to exec into it directly from the browser.
+
+<video autoplay loop muted playsinline>
+  <source src="/assets/web-shell.webm" type="video/webm">
+</video>
+
+## Keyboard Shortcuts
+
+| Shortcut | Action |
+|----------|--------|
+| `Ctrl+K` / `Cmd+K` | Open command palette |
+| `Ctrl+S` / `Cmd+S` | Save editors |
+| `Escape` | Close command palette |
+| `Arrow keys` | Navigate command list |
+| `Enter` | Execute selected command |
+
+## Starting the Server
+
+```bash
+# Default: http://0.0.0.0:8000
+cf web
+
+# Custom port
+cf web --port 3000
+
+# Development mode with auto-reload
+cf web --reload
+
+# Bind to specific interface
+cf web --host 127.0.0.1
+```
+
+## Requirements
+
+The web UI requires additional dependencies:
+
+```bash
+# If installed via pip
+pip install compose-farm[web]
+
+# If installed via uv
+uv tool install 'compose-farm[web]'
+```
+
+## Architecture
+
+The web UI uses:
+
+- **FastAPI** - Backend API and WebSocket handling
+- **HTMX** - Dynamic page updates without full reloads
+- **DaisyUI + Tailwind** - Theming and styling
+- **Monaco Editor** - Code editing for compose/env files
+- **xterm.js** - Terminal emulation for logs and shell access

examples/README.md

@@ -1,42 +1,156 @@
 # Compose Farm Examples
 
-This folder contains example Docker Compose services for testing Compose Farm locally.
+Real-world examples demonstrating compose-farm patterns for multi-host Docker deployments.
+
+## Stacks
+
+| Stack | Type | Demonstrates |
+|---------|------|--------------|
+| [traefik](traefik/) | Infrastructure | Reverse proxy, Let's Encrypt, file-provider |
+| [coredns](coredns/) | Infrastructure | Wildcard DNS for `*.local` domains |
+| [mealie](mealie/) | Single container | Traefik labels, resource limits, environment vars |
+| [uptime-kuma](uptime-kuma/) | Single container | Docker socket, user mapping, custom DNS |
+| [paperless-ngx](paperless-ngx/) | Multi-container | Redis + PostgreSQL + App stack |
+| [autokuma](autokuma/) | Multi-host | Demonstrates `all` keyword (runs on every host) |
+
+## Key Patterns
+
+### External Network
+
+All stacks connect to a shared external network for inter-service communication:
+
+```yaml
+networks:
+  mynetwork:
+    external: true
+```
+
+Create it on each host with consistent settings:
+
+```bash
+compose-farm init-network --network mynetwork --subnet 172.20.0.0/16
+```
+
+### Traefik Labels (Dual Routes)
+
+Stacks expose two routes for different access patterns:
+
+1. **HTTPS route** (`websecure` entrypoint): For your custom domain with Let's Encrypt TLS
+2. **HTTP route** (`web` entrypoint): For `.local` domains on your LAN (no TLS needed)
+
+This pattern allows accessing stacks via:
+- `https://mealie.example.com` - from anywhere, with TLS
+- `http://mealie.local` - from your local network, no TLS overhead
+
+```yaml
+labels:
+  # HTTPS route for custom domain (e.g., mealie.example.com)
+  - traefik.enable=true
+  - traefik.http.routers.myapp.rule=Host(`myapp.${DOMAIN}`)
+  - traefik.http.routers.myapp.entrypoints=websecure
+  - traefik.http.services.myapp.loadbalancer.server.port=8080
+  # HTTP route for .local domain (e.g., myapp.local)
+  - traefik.http.routers.myapp-local.rule=Host(`myapp.local`)
+  - traefik.http.routers.myapp-local.entrypoints=web
+```
+
+> **Note:** `.local` domains require local DNS to resolve to your Traefik host.
+> The [coredns](coredns/) example provides this - edit `Corefile` to set your Traefik IP.
+
+### Environment Variables
+
+Each stack has a `.env` file for secrets and domain configuration.
+Edit these files to set your domain and credentials:
+
+```bash
+# Example: set your domain
+echo "DOMAIN=example.com" > mealie/.env
+```
+
+Variables like `${DOMAIN}` are substituted at runtime by Docker Compose.
+
+### NFS Volume Mounts
+
+All data is stored on shared NFS storage at `/mnt/data/`:
+
+```yaml
+volumes:
+  - /mnt/data/myapp:/app/data
+```
+
+This allows stacks to migrate between hosts without data loss.
+
+### Multi-Host Stacks
+
+Stacks that need to run on every host (e.g., monitoring agents):
+
+```yaml
+# In compose-farm.yaml
+stacks:
+  autokuma: all  # Runs on every configured host
+```
+
+### AutoKuma Labels (Optional)
+
+The autokuma example demonstrates compose-farm's **multi-host feature** - running the same stack on all hosts using the `all` keyword. AutoKuma itself is not part of compose-farm; it's just a good example because it needs to run on every host to monitor local Docker containers.
+
+[AutoKuma](https://github.com/BigBoot/AutoKuma) automatically creates Uptime Kuma monitors from Docker labels:
+
+```yaml
+labels:
+  - kuma.myapp.http.name=My App
+  - kuma.myapp.http.url=https://myapp.${DOMAIN}
+```
 
 ## Quick Start
 
 ```bash
 cd examples
 
-# Check status of all services
+# 1. Create the shared network on all hosts
+compose-farm init-network
+
+# 2. Start infrastructure (reverse proxy + DNS)
+compose-farm up traefik coredns
+
+# 3. Start other stacks
+compose-farm up mealie uptime-kuma
+
+# 4. Check status
 compose-farm ps
 
-# Pull images
-compose-farm pull --all
+# 5. Generate Traefik file-provider config for cross-host routing
+compose-farm traefik-file --all
 
-# Start hello-world (runs and exits)
-compose-farm up hello
+# 6. View logs
+compose-farm logs mealie
 
-# Start nginx (stays running)
-compose-farm up nginx
-
-# Check nginx is running
-curl localhost:8080
-
-# View logs
-compose-farm logs nginx
-
-# Stop nginx
-compose-farm down nginx
-
-# Update all (pull + restart)
-compose-farm update --all
+# 7. Stop everything
+compose-farm down --all
 ```
 
-## Services
+## Configuration
 
-- **hello**: Simple hello-world container (exits immediately)
-- **nginx**: Nginx web server on port 8080
+The `compose-farm.yaml` shows a multi-host setup:
 
-## Config
+- **primary** (192.168.1.10): Runs Traefik and heavy stacks
+- **secondary** (192.168.1.11): Runs lighter stacks
+- **autokuma**: Runs on ALL hosts to monitor local containers
 
-The `compose-farm.yaml` in this directory configures both services to run locally (no SSH).
+When Traefik runs on `primary` and a stack runs on `secondary`, compose-farm
+automatically generates file-provider config so Traefik can route to it.
+
+## Traefik File-Provider
+
+When stacks run on different hosts than Traefik, use `traefik-file` to generate routing config:
+
+```bash
+# Generate config for all stacks
+compose-farm traefik-file --all -o traefik/dynamic.d/compose-farm.yml
+
+# Or configure auto-generation in compose-farm.yaml:
+traefik_file: /opt/stacks/traefik/dynamic.d/compose-farm.yml
+traefik_stack: traefik
+```
+
+With `traefik_file` configured, compose-farm automatically regenerates the config after `up`, `down`, and `update` commands.

examples/autokuma/.env

@@ -0,0 +1,4 @@
+# Copy to .env and fill in your values
+DOMAIN=example.com
+UPTIME_KUMA_USERNAME=admin
+UPTIME_KUMA_PASSWORD=your-uptime-kuma-password

examples/autokuma/compose.yaml

@@ -0,0 +1,31 @@
+# AutoKuma - Automatic Uptime Kuma monitor creation from Docker labels
+# Demonstrates: Multi-host service (runs on ALL hosts)
+#
+# This service monitors Docker containers on each host and automatically
+# creates Uptime Kuma monitors based on container labels.
+#
+# In compose-farm.yaml, configure as:
+#   autokuma: all
+#
+# This runs the same container on every host, so each host's local
+# Docker socket is monitored.
+name: autokuma
+services:
+  autokuma:
+    image: ghcr.io/bigboot/autokuma:latest
+    container_name: autokuma
+    restart: unless-stopped
+    environment:
+      # Connect to your Uptime Kuma instance
+      AUTOKUMA__KUMA__URL: https://uptime.${DOMAIN}
+      AUTOKUMA__KUMA__USERNAME: ${UPTIME_KUMA_USERNAME}
+      AUTOKUMA__KUMA__PASSWORD: ${UPTIME_KUMA_PASSWORD}
+      # Tag for auto-created monitors
+      AUTOKUMA__TAG__NAME: autokuma
+      AUTOKUMA__TAG__COLOR: "#10B981"
+    volumes:
+      # Access local Docker socket to discover containers
+      - /var/run/docker.sock:/var/run/docker.sock:ro
+    # Custom DNS for resolving internal domains
+    dns:
+      - 192.168.1.1  # Your local DNS server

examples/compose-farm-state.yaml

@@ -0,0 +1,10 @@
+deployed:
+  autokuma:
+  - primary
+  - secondary
+  - local
+  coredns: primary
+  mealie: secondary
+  paperless-ngx: primary
+  traefik: primary
+  uptime-kuma: secondary

examples/compose-farm.yaml

@@ -1,11 +1,41 @@
-# Example Compose Farm config for local testing
-# Run from the examples directory: cd examples && compose-farm ps
+# Example Compose Farm configuration
+# Demonstrates a multi-host setup with NFS shared storage
+#
+# To test locally: Update the host addresses and run from the examples directory
 
-compose_dir: .
+compose_dir: /opt/stacks/compose-farm/examples
+
+# Auto-regenerate Traefik file-provider config after up/down/update
+traefik_file: /opt/stacks/compose-farm/examples/traefik/dynamic.d/compose-farm.yml
+traefik_stack: traefik  # Skip Traefik's host in file-provider (docker provider handles it)
 
 hosts:
+  # Primary server - runs Traefik and most stacks
+  # Full form with all options
+  primary:
+    address: 192.168.1.10
+    user: deploy
+    port: 22
+
+  # Secondary server - runs some stacks for load distribution
+  # Short form (user defaults to current user, port defaults to 22)
+  secondary: 192.168.1.11
+
+  # Local execution (no SSH) - for testing or when running on the host itself
   local: localhost
 
-services:
-  hello: local
-  nginx: local
+stacks:
+  # Infrastructure (runs on primary where Traefik is)
+  traefik: primary
+  coredns: primary  # DNS for *.local resolution
+
+  # Multi-host stacks (runs on ALL hosts)
+  # AutoKuma monitors Docker containers on each host
+  autokuma: all
+
+  # Primary server stacks
+  paperless-ngx: primary
+
+  # Secondary server stacks (distributed for performance)
+  mealie: secondary
+  uptime-kuma: secondary

examples/coredns/.env

@@ -0,0 +1,2 @@
+# CoreDNS doesn't need environment variables
+# The Traefik IP is configured in the Corefile

examples/coredns/Corefile

@@ -0,0 +1,22 @@
+# CoreDNS configuration for .local domain resolution
+#
+# Resolves *.local to the Traefik host IP (where your reverse proxy runs).
+# All other queries are forwarded to upstream DNS.
+
+# Handle .local domains - resolve everything to Traefik's host
+local {
+    template IN A {
+        answer "{{ .Name }} 60 IN A 192.168.1.10"
+    }
+    template IN AAAA {
+        # Return empty for AAAA to avoid delays on IPv4-only networks
+        rcode NOERROR
+    }
+}
+
+# Forward everything else to upstream DNS
+. {
+    forward . 1.1.1.1 8.8.8.8
+    cache 300
+    errors
+}

examples/coredns/compose.yaml

@@ -0,0 +1,27 @@
+# CoreDNS - DNS server for .local domain resolution
+#
+# Demonstrates:
+# - Wildcard DNS for *.local domains
+# - Config file mounting from stack directory
+# - UDP/TCP port exposure
+#
+# This enables all the .local routes in the examples to work.
+# Point your devices/router DNS to this server's IP.
+name: coredns
+services:
+  coredns:
+    image: coredns/coredns:latest
+    container_name: coredns
+    restart: unless-stopped
+    networks:
+      - mynetwork
+    ports:
+      - "53:53/udp"
+      - "53:53/tcp"
+    volumes:
+      - ./Corefile:/root/Corefile:ro
+    command: -conf /root/Corefile
+
+networks:
+  mynetwork:
+    external: true

examples/hello/docker-compose.yml

@@ -1,4 +0,0 @@
-services:
-  hello:
-    image: hello-world
-    container_name: sdc-hello

examples/mealie/.env

@@ -0,0 +1,2 @@
+# Copy to .env and fill in your values
+DOMAIN=example.com

examples/mealie/compose.yaml

@@ -0,0 +1,47 @@
+# Mealie - Recipe manager
+# Simple single-container service with Traefik labels
+#
+# Demonstrates:
+# - HTTPS route: mealie.${DOMAIN} (e.g., mealie.example.com) with Let's Encrypt
+# - HTTP route: mealie.local for LAN access without TLS
+# - External network, resource limits, environment variables
+name: mealie
+services:
+  mealie:
+    image: ghcr.io/mealie-recipes/mealie:latest
+    container_name: mealie
+    restart: unless-stopped
+    networks:
+      - mynetwork
+    ports:
+      - "9925:9000"
+    deploy:
+      resources:
+        limits:
+          memory: 1000M
+    volumes:
+      - /mnt/data/mealie:/app/data
+    environment:
+      ALLOW_SIGNUP: "false"
+      PUID: 1000
+      PGID: 1000
+      TZ: America/Los_Angeles
+      MAX_WORKERS: 1
+      WEB_CONCURRENCY: 1
+      BASE_URL: https://mealie.${DOMAIN}
+    labels:
+      # HTTPS route: mealie.example.com (requires DOMAIN in .env)
+      - traefik.enable=true
+      - traefik.http.routers.mealie.rule=Host(`mealie.${DOMAIN}`)
+      - traefik.http.routers.mealie.entrypoints=websecure
+      - traefik.http.services.mealie.loadbalancer.server.port=9000
+      # HTTP route: mealie.local (for LAN access, no TLS)
+      - traefik.http.routers.mealie-local.rule=Host(`mealie.local`)
+      - traefik.http.routers.mealie-local.entrypoints=web
+      # AutoKuma: automatically create Uptime Kuma monitor
+      - kuma.mealie.http.name=Mealie
+      - kuma.mealie.http.url=https://mealie.${DOMAIN}
+
+networks:
+  mynetwork:
+    external: true

examples/nginx/docker-compose.yml

@@ -1,6 +0,0 @@
-services:
-  nginx:
-    image: nginx:alpine
-    container_name: sdc-nginx
-    ports:
-      - "8080:80"

examples/paperless-ngx/.env

@@ -0,0 +1,4 @@
+# Copy to .env and fill in your values
+DOMAIN=example.com
+POSTGRES_PASSWORD=change-me-to-a-secure-password
+PAPERLESS_SECRET_KEY=change-me-to-a-long-random-string