request-refactor-plan
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is instructional and focuses on a legitimate software development task. It contains no hardcoded credentials, malicious remote code execution patterns, or obfuscation.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data by reading codebase files and user-provided descriptions, creating a surface for potential indirect prompt injection. However, this is inherent to the skill's primary purpose and no malicious usage was found.
- Ingestion points: The agent reads the local repository state (Step 2) and checks for test coverage (Step 6), in addition to user interview responses.
- Boundary markers: None; the skill does not specify delimiters to separate untrusted data from the agent's instructions.
- Capability inventory: The skill explores the codebase and uses a GitHub integration to create issues.
- Sanitization: None; data collected from the codebase and user is directly interpolated into a GitHub issue template.
Audit Metadata