archive-searcher
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Prompt Injection (LOW): The skill documentation describes a workflow (Category 8: Indirect Prompt Injection) where archive matches are aggregated and sent to an AI model for synthesis. This pattern is vulnerable to malicious instructions embedded in the archived data.
- Ingestion points:
references/ai-integration.md(autonomous_search function). - Boundary markers: The provided pseudo-code does not demonstrate the use of delimiters or 'ignore' instructions for the AI.
- Capability inventory: Use of
subprocess.runto executeripgrepandgitacross multiple files. - Sanitization: No sanitization of ingested file content is described before it is provided as context to the LLM.
Audit Metadata