clean-code-refactoring
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute several shell commands to manage the refactoring process. These include Git operations (
git commit), formatting tools (npx prettier,ruff format,gofmt), and architectural/testing scripts (python3 tools/cc-scan.py,python3 tools/arch-scan.py,python3 tools/test-lock.py). - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided code and incorporates it into a workflow involving shell execution, posing a potential injection risk.
- Ingestion points: User-provided code snippets or file paths targetted for refactoring (e.g., "refactor this", "clean this up") as described in the metadata and Section 8 (SKILL.md).
- Boundary markers: None identified; the instructions do not specify delimiters or instructions for the agent to ignore embedded commands within the code being refactored.
- Capability inventory: The skill uses
npxandpython3to execute commands, specifically allowing a user-controlled test command via the--test-cmdflag intest-lock.py(Section 2, SKILL.md). - Sanitization: None identified; there are no instructions for the agent to validate or sanitize code content or command-line arguments derived from user input.
Audit Metadata