# Team Onboarding — {{PROJECT_DISPLAY_NAME}}

Welcome to the team! This guide gets you from clone to productive in under 10 minutes.

## Prerequisites

- [ ] Git access to {{GH_ORG_REPO}} (SSH key or HTTPS token)
{{#IF_LINEAR}}- [ ] Linear API key (ask your team lead — Settings > API > Personal API Keys)
{{/IF_LINEAR}}{{#IF_SLACK}}- [ ] Slack workspace access
{{/IF_SLACK}}
## Setup

### 1. Clone and install

```bash
git clone <repo-url>
cd {{PROJECT_NAME}}
npm install
```

### 2. Generate your worktree slots

```bash
npx traqr render
source scripts/{{PREFIX}}-aliases.sh
```

This creates your local worktree slots (feature1, bugfix1, etc.) and shell aliases.

{{#IF_MEMORY_FULL}}### 3. Connect to the shared memory DB

Your team uses a shared memory database. Set it up with the interactive wizard:

```bash
npx traqr-memory-mcp --install
```

The wizard will ask for your database and embedding provider. **Get these from your team lead:**

{{#IF_GITHUB}}- **Supabase URL**: `https://xxx.supabase.co`
- **Supabase Service Role Key**: `eyJ...` (NOT the anon key)
- **OpenAI API Key** (or Gemini): for generating embeddings
{{/IF_GITHUB}}{{#IF_GITLAB}}- **Database URL**: `postgresql://user:pass@host:5432/dbname` (Postgres connection string)
- **Embedding Provider**: Check with your team (OpenAI, Gemini, or other supported provider)
{{/IF_GITLAB}}
The wizard writes your `.claude/mcp.json` automatically. This file is gitignored (it contains secrets).

### 4. Verify memory works

Open Claude Code and run:
```
/startup
```

You should see it connect to the memory DB and load context. If it fails:
- Check your credentials in `.claude/mcp.json`
- Make sure the database is accessible from your network
- Ask your team lead for help

{{/IF_MEMORY_FULL}}### {{#IF_MEMORY_FULL}}5{{/IF_MEMORY_FULL}}{{^IF_MEMORY_FULL}}3{{/IF_MEMORY_FULL}}. Start working

```bash
z1 && claude    # Jump to slot 1, open Claude Code
```

Claude will read the project's `CLAUDE.md` and understand the full workflow automatically.

## Your First Task

{{#IF_ISSUES}}1. Pick a ticket from {{#IF_LINEAR}}Linear{{/IF_LINEAR}}{{#IF_GITLAB_ISSUES}}GitLab Issues{{/IF_GITLAB_ISSUES}}{{#IF_GITHUB_ISSUES}}GitHub Issues{{/IF_GITHUB_ISSUES}} (ask your team lead for a good starter)
{{/IF_ISSUES}}{{^IF_ISSUES}}1. Ask your team lead for a good starter task
{{/IF_ISSUES}}2. Jump into a slot: `z1 && claude`
{{#IF_MEMORY_FULL}}3. Claude runs `/startup` → `/context` → you implement → `/ship`
{{/IF_MEMORY_FULL}}{{^IF_MEMORY_FULL}}3. Implement the task → run `/ship` to create a PR
{{/IF_MEMORY_FULL}}
## Quick Reference

| Command | What it does |
|---------|-------------|
| `z1` | Jump to feature slot 1 |
| `c1` | Open Claude Code in slot 1 |
| `/ship "Title"` | Commit, push, create PR |
| `/sync` | Sync all slots with main |
{{#IF_MEMORY_FULL}}| `/startup` | Load memories, claim tickets |
| `/context "task"` | Search memory for relevant context |
{{/IF_MEMORY_FULL}}
## What's Shared vs Personal

| File | In Repo | Gitignored | Notes |
|------|:-------:|:----------:|-------|
| `.traqr/config.json` | x | | Project config (no secrets) |
| `CLAUDE.md` | x | | AI instructions (auto-generated) |
| `.traqr/scripts/` | x | | Shell aliases and hooks |
| `.claude/mcp.json` | | x | Your memory DB credentials |
| `.env.local` | | x | Your API keys |

## Need Help?

- **Read `CLAUDE.md`** — it has the full project guide
- **Ask your team lead** — they set up the project and know the config
- **Check Common Gotchas** — bottom of `CLAUDE.md` has known issues

---
*Generated by TraqrOS — Automation Score: {{AUTOMATION_SCORE}}/100*