composio-exa
Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: Communicates with backend.composio.dev to perform searches and retrieve content. This is a well-known service for AI agent integrations and is consistent with the skill's purpose.
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it retrieves content from external websites.
- Ingestion points: Action endpoints in references/actions.md (EXA_SEARCH, EXA_ANSWER, EXA_GET_CONTENTS_ACTION).
- Boundary markers: None identified in the skill definition.
- Capability inventory: Network requests to the Composio API.
- Sanitization: Content from the web is processed without explicit sanitization steps within the skill.
- [SAFE]: No malicious patterns, obfuscation, or suspicious persistence mechanisms were detected. The skill follows established patterns for Composio tool integrations.
Audit Metadata