refactor-loop

Pass

Audited by Gen Agent Trust Hub on Apr 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill is designed to automate a refactoring workflow by executing various CLI tools including unilyze, git, jq, and dotnet. These commands are used to calculate code health metrics, identify high-churn files, and execute test suites. While standard for developer tools, these commands run with the agent's privileges over the local filesystem.
  • [PROMPT_INJECTION]: The skill processes external, potentially untrusted data which creates an indirect prompt injection surface (Category 8).
  • Ingestion points: The skill reads project source code and JSON-formatted metrics from files like quality-audit.json and refactor-before.json into the agent's context.
  • Boundary markers: There are no explicit instructions or delimiters defined to separate the processed source code or JSON data from the agent's internal instructions.
  • Capability inventory: The skill allows for shell command execution (unilyze, dotnet test, git) and file system modifications (writing to .unilyze/ and refactoring code).
  • Sanitization: No sanitization or verification steps are specified for the content of the source code or the metric data before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 9, 2026, 05:40 PM
Security Audit — agent-trust-hub — refactor-loop