hindsight-docs
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides documentation for the Hindsight system. It contains no executable scripts, tool definitions, or background processes. Navigational instructions guide the agent to use standard system tools (Read, Grep, Glob) for informational purposes only.
- [PROMPT_INJECTION]: A potential injection pattern was flagged in 'references/sdks/integrations/superagent.md'. Technical review confirmed this text is inside a string literal within a Python code block, serving as an example for a security feature that detects prompt injections. It is not an instruction to the AI agent.
Audit Metadata