skills/evan-acg/skills/dev-flow/Gen Agent Trust Hub

dev-flow

Pass

Audited by Gen Agent Trust Hub on Oct 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The quality gate scripts in the Python (templates/python-quality/scripts/crap.py) and Node.js (templates/node-vitest-quality/scripts/quality/mutation.mjs) templates execute external commands such as radon, git, and stryker. These executions are performed using standard library functions (subprocess.run and execFileSync) and are restricted to project-specific development tools as part of the intended quality verification process.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze project source code and coverage reports during the verification lifecycle. While this represents a potential attack surface where malicious instructions could be embedded in source code comments or metadata, the workflow mitigates this by prioritizing deterministic quality tools (linters, complexity analyzers) over LLM-based interpretation of untrusted data.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 1, 2026, 03:21 PM