skill-health-audit

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill is purely instructional and does not contain any executable scripts (Python, JavaScript, or Shell). It serves as a framework for the AI agent to perform manual audits of other skills based on provided guidelines.- [SAFE]: The skill explicitly prohibits autonomous actions. The instructions in SKILL.md and SKILL_CARD.md state that the audit must not edit files, promote drafts, or act as an automatic editor, ensuring human-in-the-loop control.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process third-party skill files which could contain malicious instructions. However, the risk is minimal as the skill is limited to generating a text-based report and has no tools or permissions to execute code or write to the filesystem.
  • Ingestion points: Reads skill files, cards, metadata, and reports (referenced in SKILL_CARD.md).
  • Boundary markers: Absent in the provided text, but instructions emphasize deterministic checks and reporting over behavioral obedience.
  • Capability inventory: No file-write, network, or execution capabilities are defined or used.
  • Sanitization: Not applicable as the skill only produces a text report for a human maintainer.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 08:38 PM
Security Audit — agent-trust-hub — skill-health-audit