The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes node .agents/scripts/precheck.mjs to validate the project state and category before proceeding with documentation tasks.
[COMMAND_EXECUTION]: The skill uses git status --porcelain, git diff --name-only, and git diff --cached --name-only to identify changes in the working directory and staging area.
[INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from repository file changes and user-specified paths to update internal documentation.
Ingestion points: File paths and content changes retrieved via git commands (diff/status) and direct user input are used as the primary source for documentation alignment.
Boundary markers: None are defined; the skill reads external content directly into the agent's context to determine documentation updates.
Capability inventory: The skill has read/write access to markdown files in the .agents/docs/ directory and modifies the AGENTS.md index file.
Sanitization: There is no evidence of sanitization, filtering, or instruction-override protection when processing data from the repository's file system.

sync-docs