clean-code
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- PROMPT_INJECTION (LOW): Susceptible to indirect prompt injection (Category 8) as it reads full contents of untrusted code files. Evidence Chain: (1) Ingestion: resolve targets in Step 1 and read contents in Step 3. (2) Boundary: None specified. (3) Capability: File modification in Step 4 and command execution in Step 5. (4) Sanitization: No content validation.
- COMMAND_EXECUTION (LOW): Instructs the agent to run project-defined linters and tests in Step 5, which could lead to command injection if project configurations are malicious.
- NO_CODE (SAFE): The skill consists only of markdown instructions with no accompanying executable scripts or binaries.
Audit Metadata