test-tagging

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data in the form of existing test source code, which could contain malicious instructions designed to mislead the agent.\n
  • Ingestion points: The skill reads test project folders and specific test files provided in the 'Test project or files' input.\n
  • Boundary markers: The instructions do not specify using delimiters or explicit 'ignore instructions' warnings when processing test bodies.\n
  • Capability inventory: The skill can modify source files ('auto-edit' capability in SKILL.md) and execute shell commands for verification ('dotnet build', 'pytest', 'mvn test-compile', 'go vet', etc.).\n
  • Sanitization: No explicit sanitization or validation of the test file content is described beyond identifying test frameworks.\n- [COMMAND_EXECUTION]: The skill invokes local build and test-collection tools to verify that source code modifications did not break the project.\n
  • Evidence: The 'Verify edits before reporting' section (Step 6) details the use of dotnet build, pytest --collect-only, mvn test-compile, go vet ./..., cargo check --tests, and npm run test:list.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 07:35 AM
Security Audit — agent-trust-hub — test-tagging