Skip to content
Back to skills

Human In The Loop

ASecurity

Load the Law 37 human-gates playbook — gate tiers from the Law 2 severity (Trivial / Standard / Significant), the committed intent.md → spec.md → plan.md artifact chain in docs/features/<id>/ with owner approval rules, the PR Intake block and Decision log, risk-tiered review (Claude Code's /code-review on Standard and Significant work, a fresh reviewer subagent with engineering:code-review on Significant work, /code-review ultra offered in the merge order for auth, data, migrations and the ho...

  • 4 stars
  • 0 votes
  • 0 copies
  • 3 views
  • Added October 8, 2026
developmentcode-review

Works with

  • claude code

Security analysis

A100/100

Scanned October 8, 2026

npx -y skills add BojanKocijan/bk-charterline --skill human-in-the-loop --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Human In The Loop?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Human In The Loop
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/bojankocijan-human-in-the-loop-bk-charterline/badge)](https://www.skillsdirectory.com/skills/bojankocijan-human-in-the-loop-bk-charterline)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: human-in-the-loop
description: Load the Law 37 human-gates playbook — gate tiers from the Law 2 severity (Trivial / Standard / Significant), the committed intent.md → spec.md → plan.md artifact chain in docs/features/<id>/ with owner approval rules, the PR Intake block and Decision log, risk-tiered review (Claude Code's /code-review on Standard and Significant work, a fresh reviewer subagent with engineering:code-review on Significant work, /code-review ultra offered in the merge order for auth, data, migrations and the hook), the review cap on AI-authored PRs awaiting review, the review queue digest, and the delegable vs judgment-heavy triage. Auto-load before any Medium-or-High change, before drafting any non-chore PR, and when the user says "review queue", "approve intent", "approve spec", "approve plan", "review cap", or "skip gates".
license: GPL-3.0-only
---

# Human in the loop — Law 37

Full reference: `Read knowledge/HUMAN_IN_THE_LOOP.md`. Load it before acting; this skill is the trigger and the summary.

## When to invoke

- The Law 2 announcement is Medium or High severity, or the change is a new feature or user flow
- Drafting any non-`chore:` PR (Intake block + Decision log)
- `review queue`, `approve intent` / `approve spec` / `approve plan`, `review cap <N>` / `review cap off`, `skip gates`

## The five rules in one screen

| Rule | What Claude does |
|---|---|
| Gate tier | States `**Gate tier:**` in the announcement. Trivial: nothing extra. Standard: approved `plan.md`. Significant: `intent.md` → `spec.md` → `plan.md`, approved in order. |
| Committed artifacts | Writes them in `docs/features/<id>/`, links existing stories or design files instead of duplicating, stops after each one, records `Approved-by` only after explicit approval. |
| PR explains itself | Intake block + Decision log under `## Summary`; size per Law 31; names every touched test. |
| Review follows risk | Standard: diff checked against the plan. Significant: independent subagent review, plus a second named reviewer on a team. AI review is never the merge decision. |
| Review capacity | Cap of 3 AI PRs awaiting the user's review; sorts work into delegable vs judgment-heavy; batches ready PRs into one `review queue` digest. |

## Never

- Code before an approved `plan.md` on Standard or Significant work
- An `Approved-by` line without the owner's explicit approval in this session
- Lowering a gate tier without the user's `skip gates`
- Treating any AI review, including Claude's own, as approval to merge

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.