macos_security/src/content/docs/api/classes/baseline.md

title, description, sidebar

title	description	sidebar
mscp.classes.baseline	Baseline document model.	order 1

Source: src/mscp/classes/baseline.py

Baseline document model.

A baseline is a top-level mSCP document that pairs metadata (authors, title, description, target platform) with a set of Profile sections, each of which groups one or more Macsecurityrule objects. This module defines the Baseline, Profile, and Author Pydantic models, along with class methods to load baselines from YAML and write them back out.

Re-exports (__all__)

Author, Profile, Baseline

Classes

Author

class Author(BaseModelWithAccessors)

One author or owning organisation of a baseline.

Attributes

• name (str | None) — Personal name of the author, if available.
• organization (str | None) — Organisation the author represents, if applicable.

Profile

class Profile(BaseModelWithAccessors)

A named section of a baseline grouping related rules.

Profiles correspond to the top-level groupings rendered in generated guidance (Auditing, Authentication, Operating System, etc.), plus the synthetic special sections (Inherent, Permanent, Not Applicable, Supplemental).

Attributes

• section (str) — Display name of the section (e.g. "Operating System").
• description (str) — Section description copied from the matching section YAML file.
• rules (list[Macsecurityrule]) — Rules included in this profile, generally sorted by rule_id.

Baseline

class Baseline(BaseModelWithAccessors)

An mSCP baseline document.

A baseline pairs metadata about a security guide (title, description, authors, target platform) with the Profile sections that hold its rules. Instances are normally constructed via from_yaml (loading an existing baseline file) or create_new (assembling one from a rule set).

Attributes

• authors (list[Author]) — Authors and/or owning organisations.
• profile (list[Profile]) — Section profiles holding the baseline's rules.
• name (str) — Short identifier, typically derived from the baseline filename stem.
• title (str) — Human-readable full title of the baseline.
• description (str) — Description rendered in generated guidance.
• platform (dict[str, Any]) — Target platform metadata, e.g. {"os": "macOS", "version": 15.0}.
• parent_values (str) — Name of the parent benchmark this baseline inherits from (e.g. "recommended"), or empty.

Class Methods

from_yaml

from_yaml(cls, file_path: Path, language: str='en', custom: bool=False) -> 'Baseline'

Load a Baseline from a YAML file with rules resolved.

Reads the baseline document, then for each profile entry resolves the matching section file (under config["sections_dir"], plus config["custom"]["sections_dir"] if custom is set) and loads its rules via Macsecurityrule.load_rules.

Args

• file_path (Path) — Path to the baseline YAML file.
• language (str) — Language code passed through to the file loader for localised strings. Defaults to "en".
• custom (bool) — If true, also search the configured custom sections directory when resolving section files. Defaults to False.

Returns

• Baseline — A fully populated baseline with all profiles and rules resolved. Profiles whose section file cannot be found are skipped with a warning.

create_new

create_new(cls, output_file: Path, rules: list[Macsecurityrule], baseline_name: str | None, authors: list[Author], full_title: str, benchmark: str, os_type: str, os_version: float, baseline_dict: dict[str, Any], language: str='en') -> None

Build a new baseline from a rule set and write it to YAML.

Groups rules into profiles by their section (or by special section tags inherent, permanent, n_a, supplemental when present), loads section descriptions from config["sections_dir"], and serialises the result via to_yaml. If baseline_dict lacks a title or description they're synthesised from the other arguments.

Args

• output_file (Path) — Destination YAML file for the new baseline.
• rules (list[Macsecurityrule]) — Rules to include.
• baseline_name (str | None) — Short identifier folded into the synthesised title/description if those aren't supplied.
• authors (list[Author]) — Authors attributed to the baseline.
• full_title (str) — Long-form title prepended to the synthesised title and description when set.
• benchmark (str) — Benchmark identifier (e.g. "recommended"); stored as parent_values and used to extend the synthesised description when equal to "recommended".
• os_type (str) — Operating-system family (e.g. "macOS").
• os_version (float) — Operating-system version (e.g. 15.0).
• baseline_dict (dict[str, Any]) — Additional baseline metadata merged into the constructor; title and description are filled in if absent.
• language (str) — Language code for loaded section descriptions. Defaults to "en".

Side Effects

• Writes the generated baseline to output_file. Reads every
• *.y*ml file in config["sections_dir"] to resolve
• section descriptions.

Methods

to_dataframe

to_dataframe(self) -> pd.DataFrame

Flatten the baseline's rules into a pandas.DataFrame.

Each rule contributes one row. The nested references mapping is unpacked so each reference namespace (nist, disa, etc.) becomes its own column.

Returns

• pd.DataFrame — One row per rule across all profiles, with rule fields and unpacked references as columns.

to_yaml

to_yaml(self, output_path: Path) -> None

Serialise this baseline to YAML in canonical key order.

The serialised document orders top-level keys as title, description, authors, parent_values, platform, profile (any other keys are dropped), and orders profiles by a fixed sequence (Auditing, Authentication, iCloud, Operating System, Password Policy, System Settings, followed by the special sections). Within each profile, rules is reduced to a sorted list of rule IDs.

Args

• output_path (Path) — Destination YAML file.