compliant-phase

Pass

Audited by Gen Agent Trust Hub on Jul 24, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it processes untrusted data from external sources during its enrichment phase.
  • Ingestion points: External data enters the agent context via web_search and exa.api results in Phase 2.
  • Boundary markers: The skill definition lacks explicit delimiters or instructions for the agent to ignore potentially malicious instructions embedded within external search results.
  • Capability inventory: The skill is marked with mutating: true, which allows the agent to modify its state or environment based on the processed external data.
  • Sanitization: There is no evidence of content filtering, validation, or escaping of the data retrieved from external lookups.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 24, 2026, 02:50 PM
Security Audit — agent-trust-hub — compliant-phase