The Agent Skills Directory

[SAFE]: The skill specifies a logical framework for project risk assessment and documentation review. No evidence of malicious code execution, credential exfiltration, or persistence was found.
[PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by instructing the agent to ingest content from various repository files including README.md, brain/index.md, and docs/. There are no explicit boundary markers or sanitization procedures described to prevent the agent from following instructions embedded in those files. The agent has the capability to read local files for repo analysis and to write to .context/implementation-notes.md.

blindspot-pass