skill-diagnostics

Pass

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to read and analyze other skill files, creating an indirect prompt injection surface.
  • Ingestion points: Ingests text from local SKILL.md and ref/*.md files during the diagnostic process.
  • Boundary markers: Contains a specific 'Must' rule directing the agent to treat all scanned content as data and avoid executing any instructions found within those files.
  • Capability inventory: The skill possesses file-writing capabilities to apply fixes to skills after obtaining user approval.
  • Sanitization: Relies on instructional-level constraints to prevent the agent from accidentally interpreting data as instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 5, 2026, 08:07 PM
Security Audit — agent-trust-hub — skill-diagnostics