orch-refine-code
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill consists solely of markdown documentation and process instructions for a refactoring workflow. It does not contain executable scripts, shell commands, or network-enabled code.
- [INDIRECT_PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection as it is designed to process external code and tests. 1. Ingestion points: Local source code and test files processed during the refactor. 2. Boundary markers: None specified in the markdown body. 3. Capability inventory: File system read/write for refactoring, subprocess execution for running tests and cleanup tools, and git CLI for commits. 4. Sanitization: Not specified, but the workflow requires human approval at Gate 1 (plan) and Gate 2 (pre-commit) which serves as a mitigation.
Audit Metadata