dev-flow

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a robust TDD-based development pipeline with explicit safety gates and recovery procedures. It includes helper scripts that manage git tags and branches for checkpoints, which are standard development practices.
  • [COMMAND_EXECUTION]: The skill uses local shell scripts for environment management (e.g., safety-snapshot.sh, orphan-worktree-check.sh). These scripts perform standard git operations (tagging, branching, logging) and do not involve remote execution, credential access, or privilege escalation.
  • [DATA_EXFILTRATION]: No network operations or non-whitelisted domain requests were detected. All operations are local to the repository's git structure.
  • [PROMPT_INJECTION]: The skill contains 'HARD RULES' and 'Anti-Patterns' intended to guide the AI's behavior toward safer development practices (e.g., prohibiting destructive git commands without approval). These are not malicious overrides but rather safety constraints.
  • [EXTERNAL_DOWNLOADS]: The skill does not perform external downloads. It relies on the project's existing local test runners (e.g., Jest, Pytest) and standard toolchains.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 04:35 PM
Security Audit — agent-trust-hub — dev-flow