exa-multi-env-setup
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides search capabilities that retrieve external content from the web via the Exa API, which is then processed by the agent.
- Ingestion points: Functions in
lib/exa-search.tsandlib/exa-cache.tsretrieve content from the Exa API usingexa.searchAndContents. - Boundary markers: The provided code and instructions do not include delimiters or specific instructions to the agent to disregard potential commands or malicious patterns embedded in the external search results.
- Capability inventory: The skill demonstrates network access capabilities, communicating with both the Exa API and a Redis instance (via
ioredis) for caching. - Sanitization: There is no evidence of sanitization, filtering, or escaping of the external text and highlights retrieved from the Exa API before they are returned to the agent context.
Audit Metadata