install-loop — Claude Code skill: setup and usage
install-loop is a Claude Code skill that sets up Loop Engineering in a project through one CLI entry point. It selects a loop pattern for the task and keeps the first week report-only.
MIT ★ 11,819 updated 2026-10-29
/plugin marketplace add cobusgreyling/loop-engineering
/plugin install loop-engineering@loop-engineeringWhat install-loop does
The skill configures Loop Engineering in a repository. It picks a pattern and tool, then scaffolds the project with npx @cobusgreyling/loop. Patterns cover daily triage, dependency cleanup and post-merge TODOs. The skill runs doctor and reports the actions it prints.
When LOOP.md or STATE.md already exists, it runs doctor and status instead of scaffolding again, unless asked to rebuild. The first week is report-only: auto-fixes and auto-merges stay off. The CLI runs through npx @cobusgreyling/loop.
What it helps with
- scaffold Loop Engineering in a repository
- choose a pattern for CI, dependency or issue work
- check an existing Loop setup
- get next actions from the project health check
Example prompts for the agent
- Set up Loop Engineering in this project for daily triage.
- Configure a dependency cleanup loop and run
doctor. LOOP.mdalready exists; check the Loop status without scaffolding again.
How to install install-loop
1Plugin marketplace
/plugin marketplace add cobusgreyling/loop-engineering/plugin install loop-engineering@loop-engineeringThis installs the whole loop-engineering plugin with all its skills
2Skills CLI (Vercel)
npx skills add cobusgreyling/loop-engineering --skill install-loop -a claude-code3Manually: put the files in place
~/.claude/skills/install-loop/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/install-loop/
1Skills CLI (Vercel)
npx skills add cobusgreyling/loop-engineering --skill install-loop -a codex2Manually: put the files in place
.agents/skills/install-loop/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add cobusgreyling/loop-engineering --skill install-loop -a cursor2Manually: put the files in place
.cursor/skills/install-loop/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install cobusgreyling/loop-engineering skills/install-loop2Manually: put the files in place
.github/skills/install-loop/SKILL.md1Skills CLI (Vercel)
npx skills add cobusgreyling/loop-engineering --skill install-loop -a gemini-cli2Platform CLI
gemini skills install https://github.com/cobusgreyling/loop-engineering --path skills/install-loopManually: put the files in place
.devin/skills/install-loop/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about install-loop
How do I install install-loop in Claude Code?
Run /plugin marketplace add cobusgreyling/loop-engineering, then /plugin install loop-engineering@loop-engineering. Another installation method is npx skills add cobusgreyling/loop-engineering --skill install-loop -a claude-code.
What does install-loop do?
It configures Loop Engineering in a project, selects a pattern and tool, then runs doctor. The first week stays report-only.
What does install-loop need to run?
It invokes the CLI with npx @cobusgreyling/loop. The skill can be installed in Claude Code through the marketplace commands or with npx skills.
Source text
SKILL.md · MIT
---
name: install-loop
description: >
Install Loop Engineering into a project via the unified CLI front door
(@cobusgreyling/loop). Prefer this over invoking loop-init / loop-audit
separately. Week-one is report-only; never enable auto-merge or unattended
fixes unless the human explicitly asks and doctor is healthy.
---
# Install Loop Engineering
## Goal
Leave the repo with a scaffolded **report-only** loop, a **Loop Ready** score, and clear next actions — using the **single** CLI:
```bash
npx @cobusgreyling/loop
```
Do **not** deprecate or remove existing `loop-init` / `loop-audit` usages you find; both doors stay valid. Prefer `loop` for new work.
## Steps
1. **Detect context**
- If `LOOP.md` or `STATE.md` already exists, run doctor only (skip init unless human asks to re-scaffold):
```bash
npx @cobusgreyling/loop doctor . --json
npx @cobusgreyling/loop status .
```
- Otherwise continue.
2. **Pick pattern** (default: `daily-triage`)
| Pain | Pattern |
|------|---------|
| Morning chaos / unclear priorities | `daily-triage` |
| PRs stalling | `pr-babysitter` |
| CI red / flakes | `ci-sweeper` |
| CVE / Dependabot noise | `dependency-sweeper` |
| Post-merge TODOs | `post-merge-cleanup` |
| Stale release notes | `changelog-drafter` |
| Noisy issues | `issue-triage` |
| One GitHub Action, almost no files | `thin-loop` |
3. **Pick tool** — `claude` | `grok` | `codex` | `opencode` (default `claude`).
4. **Scaffold**
```bash
npx @cobusgreyling/loop init . --pattern <pattern> --tool <tool>
```
Optional harness (only if human wants Foundry): add `--with-foundry`.
5. **Doctor**
```bash
npx @cobusgreyling/loop doctor .
```
- Exit `0` = healthy · `1` = warnings · `2` = blocked
- Follow the **top 3** printed actions; do not invent extra architecture.
6. **Cost check** before high-cadence schedules:
```bash
npx @cobusgreyling/loop cost -p <pattern> -l L1 -c 1d
```
7. **Week-one rule**
- Report only. No auto-fix, no auto-merge.
- Tell the human the first `/loop` or scheduler command from init output.
- Suggest badge only when score is strong: `npx @cobusgreyling/loop badge .`
## Compatibility
These remain fully supported (do not rewrite existing docs/scripts unless asked):
```bash
npx @cobusgreyling/loop-init .
npx @cobusgreyling/loop-audit . --suggest
```
## Stop conditions
- Stop after doctor + first-run instructions unless the human asks for L2+.
- Never expand into fleet/memory/foundry unless score ≥ 80 **and** human opts in.
© Cobus Greyling · MIT · cobusgreyling/loop-engineering