The Agent Skills Directory

[SAFE]: The skill's primary function is document transformation and project maintenance. It operates entirely on local project files and includes automated verification steps to ensure integrity.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from existing documentation files to generate new configuration files, which is a standard requirement for migration tools. The risk is minimized by the skill's design and user-confirmation checkpoints.
Ingestion points: Reads local project files including CLAUDE.md, AGENTS.md, and files within the MEMORY/ directory.
Boundary markers: Uses specific Markdown header logic and blockquote sections to delimit content.
Capability inventory: Performs file read/write operations and executes bundled Python scripts for validation.
Sanitization: Employs yaml.safe_load() in its coverage script to securely parse YAML frontmatter metadata.
[COMMAND_EXECUTION]: The workflow involves running local Python scripts (precheck.py and coverage.py) that are provided within the skill package. These scripts perform benign file system checks and token-based text comparison without utilizing dangerous shell functions or external downloads.

yzr-multi-agent-context