skills/rheinmir/setup/qc-code/Gen Agent Trust Hub

qc-code

Pass

Audited by Gen Agent Trust Hub on Jul 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local commands to facilitate code review and testing. It calls python3 harness/scripts/qc-regression.py --run and project-specific test runners such as pytest, vitest, or jest.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection. It ingests untrusted code from git diff or user-specified files and uses that content to generate executable test cases (qc-*.py, qc-*.test.js).
  • Ingestion points: Reads code changes via git diff <base>..HEAD --stat and direct file reads.
  • Boundary markers: None identified. The skill does not use delimiters or instructions to ignore embedded commands within the analyzed code.
  • Capability inventory: The skill has file-write capabilities (creating test files) and command execution capabilities (running test suites).
  • Sanitization: No sanitization or validation of the code being reviewed is performed before it is used to generate logic tests.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 18, 2026, 04:03 AM
Security Audit — agent-trust-hub — qc-code