sc-improve
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes and processes external codebase directories, which serves as an ingestion point for untrusted data that could contain malicious instructions designed to influence the agent's behavior during refactoring. 1. Ingestion points: External files in directories specified by the user (e.g., 'src/', 'api-endpoints'). 2. Boundary markers: The instructions do not define specific delimiters or 'ignore' commands to isolate analyzed code from agent instructions. 3. Capability inventory: The skill can edit files, run local scripts, and transmit data to Slack, Jira, and Notion via MCP tools. 4. Sanitization: No sanitization or escaping of the ingested code content is described before it enters the agent's context.
- [COMMAND_EXECUTION]: The skill invokes local Python scripts to perform specific validation tasks. Evidence: The skill explicitly calls 'python .claude/skills/sc-principles/scripts/validate_kiss.py' and 'python .claude/skills/sc-principles/scripts/validate_purity.py' for quality and purity checks.
Audit Metadata