
Compound Engineering Setup
Checks your Compound Engineering health and repo-local config, and can scaffold Compound Packs.
- Repository
- EveryInc/compound-engineering-plugin
- Stars
- 25,850
- Licence
- MIT
- Files
- 12 files
What it does
This skill lets an AI coding agent inspect the setup of the Compound Engineering plugin in your project. It runs a bundled health-check script, reports missing or broken pieces, and offers to fix repo-local config files such as a config example, optional config.yaml, and .gitignore entries. When you ask for a pack (pack:<id>), it can scaffold a new Compound Pack and its config entry, but only after you approve.
Great for
- Developers who have installed the Compound Engineering plugin and want to confirm it is working in a repository
- Teams setting up shared repo-local configuration for Compound Engineering artifacts
- Users who want to scaffold a Compound Pack in a repo
Skip it if
- People who have not installed Compound Engineering: the health check only makes sense inside that plugin's environment
- Repositories where you do not want the agent writing config or .gitignore changes, since Phase 2 edits repo files after you approve each offer
- Anyone expecting automatic installation of optional tools: the skill explicitly reports missing tools rather than installing them
Try asking
- “Check my Compound Engineering setup and tell me what is missing”
- “Scaffold a Compound Pack called pack:research-notes in this repo”
- “Fix any invalid docs_root or config problems in this project”
- “Show me where Compound Engineering artifacts will be written”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 12 original files, the MIT licence and an attribution note.
ce-setup.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip ce-setup.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip ce-setup.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Approve or decline each repo-local change offer individually; the skill is written to ask rather than auto-configure
- Run it in a repo with a writable checkout, otherwise Phase 2 is skipped
- Review the scaffold plan before approving a new pack, since git sources in packs config are fetched from the internet
What's inside
- agents/openai.yaml
- assets/compounding-directive.md
- assets/noslop-directive.md
- assets/pack-rule-template.md
- references/acpx-cache-warm.md
- references/config-template.yaml
- references/legacy-codex-tool-map.md
- references/pack-scaffold.md
- references/project-docs.md
- references/repo-fixes.md
- scripts/packs-resolve.py
- SKILL.md
Redistributed unmodified from EveryInc/compound-engineering-plugin under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.