ponytail-audit

Pass

Audited by Gen Agent Trust Hub on Sep 26, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill creates a vulnerability surface by directing the agent to process untrusted data from an entire repository.\n
  • Ingestion points: The instructions require the agent to "Scan the whole tree" and process the "entire codebase" (SKILL.md).\n
  • Boundary markers: There are no instructions or delimiters provided to ensure the agent ignores or separates embedded instructions found within the code being audited.\n
  • Capability inventory: Risk is mitigated by the skill's restriction to a "One-shot report" and the explicit instruction that it "does not apply fixes" (SKILL.md).\n
  • Sanitization: No validation or sanitization of the codebase content is specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 26, 2026, 11:53 PM
Security Audit — agent-trust-hub — ponytail-audit