tdd-refactor
Pass
Audited by Gen Agent Trust Hub on May 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The instructions promote standard safe development practices such as small refactoring increments, vertical TDD, and ensuring backward compatibility with public APIs and schemas. No malicious intent or suspicious patterns like obfuscation or unauthorized data access were found.
- [PROMPT_INJECTION]: The skill is designed to ingest repository code and utilize the agent's ability to run shell commands for testing. While this is the intended functional purpose, it constitutes a surface for indirect prompt injection. Ingestion points: Repository instructions and nearby code/tests in SKILL.md; Boundary markers: None; Capability inventory: File editing and shell command execution for test validation in SKILL.md; Sanitization: None.
Audit Metadata