skill-stocktake

Pass

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests content from other SKILL.md files for evaluation, creating a surface for indirect prompt injection where a malicious target skill could attempt to influence the audit outcome.
  • Ingestion points: Recursive reading of skill files in the .agent/skills directory.
  • Boundary markers: None mentioned.
  • Capability inventory: Recommends file system deletions and performs web searches for package verification.
  • Sanitization: Relies on structured rubric evaluation.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 7, 2026, 04:23 PM
Security Audit — agent-trust-hub — skill-stocktake