The Agent Skills Directory

[SAFE]: No security vulnerabilities, malicious patterns, or obfuscation were identified. The skill acts as a documentation generator with clear, limited functionality.- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from the repository and conversation context. 1. Ingestion points: Project repository files and current conversation history. 2. Boundary markers: The skill uses template tags () to structure its output. 3. Capability inventory: The skill is designed to publish content to an external project issue tracker. 4. Sanitization: The instructions mandate a human review step ("Check with the user that these seams match their expectations") before the final specification is published, which serves as a mitigation for potential injection within the processed text.

to-spec