Plugin

blue-collar-governance

Deterministic AI governance runtime for Claude Code. 22 JavaScript engines compute governance decisions before the AI agent sees them — no LLM involvement in governance logic. 19 lifecycle event handlers covering prompt intake, tool execution, subagent verification, compaction integrity, config protection, and session teardown. Includes post-session verification (Foreman's Walk), append-only forensic evidence chain, three-tier autonomy control (FULL_AUTO / SUPERVISED / HARD_STOP), and operator-facing skills for governance visibility. 388 golden tests. Zero external dependencies. Fail-closed architecture: exit code 2 blocks, not prompts. Built for operators who need governance they can trust, not governance they have to hope Claude follows.

• Published source
• Repository

Analysed

11 October 2026

Last observed

12 October 2026

What the code does

• Evaluates code at runtime Code that builds and runs code while it executes.
• Runs hooks automatically Hooks that execute without being called explicitly.

Having a capability is not wrongdoing; many tools need these to do their job. It tells you what to check before you install.

This is public metadata, not a safety certification or complete assessment. Static analysis cannot establish what a tool does at runtime.