exa-search

Warn

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The installation instructions (npx skills add -g BenedictKing/exa-search) and GitHub links reference a third-party repository that does not align with the stated author ('ncdevshiv'). This discrepancy in provenance complicates the verification of the code's origin and maintenance.\n- [PROMPT_INJECTION]: The skill ingests untrusted external content from the Exa API, creating an indirect prompt injection surface where instructions hidden in search results could influence agent behavior.\n
  • Ingestion points: External web content and semantic search results retrieved via Exa API.\n
  • Boundary markers: No delimiters or safety instructions are defined to separate external data from system instructions.\n
  • Capability inventory: The skill allows for automated search and research, which could be exploited by malicious data to perform unintended actions.\n
  • Sanitization: No sanitization or validation of the retrieved content is documented.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 20, 2026, 03:28 PM
Security Audit — agent-trust-hub — exa-search