Installation¶

learnship installs as a set of workflow files into your AI platform's configuration directory. No daemon, no build step: just markdown files your agent reads.
Requirements¶
| Requirement | Notes |
|---|---|
| Node.js ≥ 18 | Required by the installer and at workflow runtime. |
| Git | Used by workflows to commit changes. |
Both are standard tools you almost certainly have already.
Quick install¶
The installer auto-detects which platforms are configured on your machine and asks whether to install globally (all projects) or locally (current project only).
Platform-specific install¶
Installs to ~/.windsurf/: workflows available as /slash-commands in all Windsurf projects. Native @agentic-learning and @impeccable skills included.
Installs to ~/.claude/: invoke as /learnship:ls, /learnship:new-project, etc.
Installs to ~/.config/opencode/: invoke as /learnship-ls, /learnship-new-project, etc.
Installs to ~/.gemini/: invoke as /learnship:ls, /learnship:new-project, etc.
Installs to ~/.codex/: invoke as $learnship-ls, $learnship-new-project, etc.
Global vs local¶
| Flag | Where it installs | Who sees it |
|---|---|---|
--global | Platform config dir (~/.windsurf/, ~/.claude/, etc.) | All your projects |
--local | Current project's config dir (.windsurf/, .claude/, etc.) | This project only |
Recommendation
Use --global for your primary platform. Use --local when you want a specific version pinned to one project or when working in a shared repo where you don't want to affect teammates.
What gets installed¶
~/.windsurf/ (Windsurf example)
├── workflows/ ← 42 workflow markdown files
│ ├── ls.md
│ ├── new-project.md
│ ├── execute-phase.md
│ └── … 39 more
└── skills/
├── agentic-learning/ ← Learning partner skill (native @invoke on Windsurf)
└── impeccable/ ← Design system skill (21 sub-skills)
For non-Windsurf platforms, skills are installed as context files:
~/.claude/learnship/
├── workflows/ ← same 42 workflows
└── skills/
├── agentic-learning/ ← loaded as AI context
└── impeccable/ ← loaded as AI context
Verify the install¶
After installing, open your AI agent and run:
You should see either a project status panel (if you're in an existing project) or a welcome message offering to run /new-project.
Updating¶
Or from inside your AI agent: /update
Local customizations
If you've edited any workflow files locally, run /reapply-patches after updating to restore your changes.