revise-skill-from-review

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill is strictly instructional and does not include any executable code, scripts, or use of platform tools that would grant network or sensitive file system access.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because its primary function is to process and edit draft content from other skills and external lists.
  • Ingestion points: The skill reads content from files in .skill-proposals/ and external 'remedy' checklists provided during the session.
  • Boundary markers: Absent; the skill relies on natural language instructions to guide the agent rather than programmatic delimiters.
  • Capability inventory: The agent is restricted to file-write operations within the .skill-proposals/ path and is explicitly forbidden from modifying production directories, committing changes, or performing installations.
  • Sanitization: No explicit sanitization of input content is performed, but the workflow includes a mandatory human review checkpoint before any revised content can be moved to the production environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 04:48 PM
Security Audit — agent-trust-hub — revise-skill-from-review