The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill operates by ingesting project data, including source code changes, dependency manifests, and configuration files, to generate documentation updates. This creates a surface where malicious content in those files could theoretically influence the agent's output.
Ingestion points: Reads package.json, .env.example, docker-compose.yml, and code diffs.
Capability inventory: Has permissions to use Write and Edit tools on the filesystem (specifically README.md).
Boundary markers: The instructions do not specify the use of delimiters or specific framing when the agent processes these external files.
Sanitization: No explicit instructions for sanitizing or escaping the content read from the project files are provided.
[SAFE]: The skill follows best practices by using placeholders (e.g., your_api_key, your_stripe_key) for sensitive information in examples. The requested tools (Read, Write, Edit, Grep) are consistent with the stated purpose of maintaining documentation.

readme-updater