Quick Start

The 30-second wedge

Drop one read-only subagent into any repo and watch it gate “done” — no wizard, no lock-in, nothing else installed:

npx -y @event4u/agent-config init --tools=claude-code --profile=minimal

Then, in Claude Code:

@production-validator check this branch is actually done

production-validator is read-only and single-purpose — it audits that no mock, stub, TODO or placeholder remains on the shipped path and that the change ran against real systems, not just green tests over hollow code.

First real task

After a full install (npx -y @event4u/agent-config init → pick your experience and tools in the wizard → Finish), start a normal task. The agent refines the request, plans it, implements, and verifies — governed by the always-active rules for your profile.

Two entry commands cover most work:

• /work "<free-form prompt>" — drive a prompt end-to-end (refine → plan → implement → test → verify), confidence-gated, no automatic git.
• /implement-ticket — the same loop, driven from a Jira/Linear ticket.

The onboarding gate

On the first turn in a project, if onboarding has not completed (onboarding.onboarded: false), the agent will ask you to run agent-config setup before anything else. Finishing the wizard flips the flag and silences the prompt. In headless/CI contexts (no .agent-settings.yml) the gate stays inert.

Next

• Profiles & Packs — pick how much governance loads.
• CLI Commands · Agent Commands