Skip to content

Getting started

Prerequisites

  • uv (recommended) or pip
  • Python 3.12+
  • Claude Code or Claude Desktop

1. Install

One-time run (no permanent install)

uvx devcoach mcp

Permanent install

uv tool install devcoach

2. Register with Claude

devcoach install

This writes the MCP server entry into both Claude Code (~/.claude.json) and Claude Desktop config. Pass --claude-code or --claude-desktop to target one only, or --force to overwrite an existing entry.

Restart Claude Code / Claude Desktop after installing.

3. Onboarding

On your first Claude session after connecting devcoach, Claude will detect that onboarding is needed and guide you through:

Option A — restore from backup

If you have a previous devcoach backup:

Do you have an existing devcoach backup to restore? Path: ~/devcoach-backup.zip

Claude calls the restore tool and your full profile (knowledge map + lessons + settings) is imported.

Option B — automatic stack detection

Claude scans your current project directory for manifest files and proposes topics:

I detected these technologies in your project:
  python        → confidence 6  (keep? or enter 0-10)
  docker        → confidence 7  (keep? or enter 0-10)
  github_actions → confidence 6  (keep? or enter 0-10)

Anything I missed?

You confirm, adjust, or skip each one. Then add anything else you work with.

Option C — manual

A free-form conversation:

Tell me about the technologies you work with day-to-day.
For each one, how confident are you?
  1–3 = still learning  ·  4–6 = comfortable  ·  7–9 = strong  ·  10 = expert

Group proposal

After your topics are finalised, Claude proposes groupings:

Here's how I'd organise these:

  Languages       → python, typescript
  Backend         → fastapi, django
  DevOps          → docker, docker_compose, github_actions
  Version Control → git, git_rebase, conventional_commits

Does this look right?

Groups are always proposed by Claude — never asked about during topic collection.

Completing setup

Claude calls complete_onboarding and saves everything. From this point on, lessons start after every technical task.

4. CLI alternative

Run the wizard in your terminal instead of through Claude:

devcoach setup

5. Your first lesson

Work on any technical task with Claude. After the response you'll see something like:

---
🎓 **devcoach** · Python · Level: Mid

**Generator expressions vs list comprehensions**

Generators are lazy: they yield one item at a time without loading everything
into memory. A list comprehension builds the entire result up front...

💡 *Senior tip:* Prefer generators any time you're chaining multiple
transformations — `sum(x*x for x in data)` never builds an intermediate list.

6. Track your progress

devcoach stats          # overview + weakest/strongest topics
devcoach lessons        # full history
devcoach profile        # knowledge map with confidence bars

Or open the web dashboard:

devcoach ui