The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill possesses an inherent attack surface because it is designed to ingest and process untrusted source code and project files, which could theoretically contain malicious instructions embedded in comments.
Ingestion points: The agent catalogs and analyzes the entire project tree, including source files (.py, .ts, etc.) and configuration files like package.json.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded prompts within the code being analyzed.
Capability inventory: The skill allows the agent to write documentation back to the filesystem, modifying existing source files or creating new ones.
Sanitization: While no automated sanitization is mentioned, the skill explicitly includes a safety guideline under 'Edge Cases' to avoid including sensitive data and to redact internal details, which acts as a manual mitigation strategy.

code-documentation