cs-refactor-ff
Pass
Audited by Gen Agent Trust Hub on Jun 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to run project-specific verification tools, such as unit tests, integration tests, type checkers, and linters, and to use
grepfor code analysis. These operations are essential for verifying refactoring results and occur entirely within the local development environment. \n- [PROMPT_INJECTION]: The skill ingests untrusted data from project files and user instructions, creating a surface for indirect prompt injection. \n - Ingestion points: Reads project configuration from
.codestable/attention.mdand processes user-provided source code files. \n - Boundary markers: Does not utilize explicit delimiters or 'ignore' instructions for the ingested content. \n
- Capability inventory: Executes local analysis and testing tools and performs file system writes for code updates. \n
- Sanitization: Relies on strict scope restrictions (modifications limited to one file and 100 lines) and automated verification steps to minimize the impact of potentially malicious data.
Audit Metadata