apollo-cost-tuning

Pass

Audited by Gen Agent Trust Hub on Mar 24, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface where external data from Apollo.io is processed by the agent.\n
  • Ingestion points: Untrusted data enters the context through the apollo.searchPeople and apollo.enrichPerson methods referenced in references/implementation-guide.md.\n
  • Boundary markers: The instructions do not define clear boundaries or provide guidance to the agent to disregard instructions potentially embedded in the fetched API data.\n
  • Capability inventory: The skill is granted significant capabilities in SKILL.md, including file system access (Write, Edit) and network operations (Bash(curl:*)), which increases the risk if malicious data were to influence agent behavior.\n
  • Sanitization: The implementation guide lacks logic for sanitizing or validating external fields (such as titles or organization data) before they are used in scoring or processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 24, 2026, 03:26 PM