explore
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill identifies repository files and external documentation as sources for analysis, which creates a surface for indirect prompt injection. This risk is inherent to the skill's primary function and is mitigated by instructions to verify claims and restrict file-writing to a documentation directory.\n
- Ingestion points: Local source code, tests, and external authoritative documentation files referenced in the workflow (SKILL.md).\n
- Boundary markers: The skill requires the agent to label claims as confirmed, inferred, or unknown and to cite specific files or symbols.\n
- Capability inventory: The agent is authorized to read repository files and write technical exploration briefs to the 'docs/agentic-engineering/explorations/' directory.\n
- Sanitization: The workflow requires the agent to distinguish documented behavior from inference, acting as a logical filter for external content.
Audit Metadata