code-search-exa
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill connects to the official Exa service endpoint at mcp.exa.ai.
- [PROMPT_INJECTION]: The skill ingests data from external sources like GitHub and StackOverflow, creating a surface for indirect prompt injection.
- Ingestion points: Results from web_search_exa (SKILL.md).
- Boundary markers: The skill uses a fork context and isolated task agents.
- Capability inventory: Limited to web search capabilities.
- Sanitization: No automated sanitization of external results is performed.
Audit Metadata