test-mutation
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill outlines a workflow where the agent processes external test reports and source code to identify and implement code changes.
- Ingestion points: Mutation testing reports (HTML/JSON) and project source files (e.g., entitlements.ts mentioned in examples).
- Boundary markers: No specific delimiters or warnings for embedded instructions are defined for the ingested data.
- Capability inventory: The skill enables shell command execution (mutation runners) and file system writes (deleting dead code, adding test assertions).
- Sanitization: The instructions do not specify sanitization or validation of the ingested reports before processing.
- [COMMAND_EXECUTION]: The skill requires the agent to execute mutation testing tools (StrykerJS, mutmut) via the command line to generate metrics and identify survivors.
- [EXTERNAL_DOWNLOADS]: The skill references industry-standard mutation testing frameworks such as StrykerJS, mutmut, and cosmic-ray. These are well-known services and do not pose an inherent security risk.
Audit Metadata