clean-code
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes instructions to execute various Python scripts for code validation, such as
ux_audit.py,security_scan.py, andlint_runner.py. These scripts are located in local directories within the environment (~/.claude/skills/) and are used to verify the quality and security of the code after modification. - [INDIRECT_PROMPT_INJECTION]: The skill defines a process for the agent to ingest and summarize output from external verification scripts. This presents a potential surface for indirect injection if a script's output contains malicious instructions. However, the skill explicitly mitigates this risk by requiring the agent to summarize the output for the user and obtain confirmation before performing any fixes ('READ → SUMMARIZE → ASK').
- Ingestion points: Shell output from verification scripts (e.g.,
security_scan.py,lint_runner.py). - Boundary markers: None explicitly defined in the data ingestion, but results are formatted into a structured markdown template for user review.
- Capability inventory: File system access (Read, Write, Edit) and local Python script execution.
- Sanitization: The skill mandates a human-in-the-loop checkpoint, preventing the agent from automatically executing instructions contained within script outputs.
- [SAFE]: The instructions regarding coding principles (SRP, DRY, KISS, YAGNI) and naming conventions are standard industry best practices aimed at improving code maintainability and readability.
Audit Metadata