refactor

Pass

Audited by Gen Agent Trust Hub on Aug 8, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill uses the codebase and filesystem tools to ingest and process external code files. This creates an attack surface where the agent could potentially encounter malicious instructions embedded in the code it is refactoring. However, as this is the primary purpose of the skill and no automated execution path for ingested code is present, the risk is minimal.
  • Ingestion points: Accesses project files via codebase and filesystem tools.
  • Boundary markers: None specified in the instructions.
  • Capability inventory: File system read and write access for refactoring tasks.
  • Sanitization: Not explicitly mentioned for ingested code content.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 8, 2026, 11:03 AM
Security Audit — agent-trust-hub — refactor