Skip to main content
Glama
README.md
# LearnWorlds MCP server for ChatGPT and Claude

LearnWorlds has an official MCP server in its initial beta on Learning Center and higher plans, checked 20 November 2026. Cotlet.dev connects ChatGPT, Claude and other AI agents to the screens in your own signed-in LearnWorlds school.

Source: [LearnWorlds integration page](https://cotlet.dev/integrations/learnworlds-api-mcp-server?utm_source=github&utm_medium=readme&utm_campaign=learnworlds).

## What can your AI agent do?

- Read learners, their tags and enrolled courses.
- Prepare enrollments and course changes for your approval.
- Review progress, completions and grades for any period.
- Read payments and product sales.
- Draft learner messages and wait for your approval before sending.

Your agent sees what your LearnWorlds login sees. Enrollments, messages and published changes wait for your OK. Payouts, billing and your LearnWorlds plan remain in your own hands.

## What can I ask?

> Enroll this week's Photography Basics learners who passed the final quiz in Portrait Lighting.

> Which courses had the most completions last month?

> Which assessments are still waiting to be graded, oldest first?

> Draft a nudge for learners who started Photography Basics over 30 days ago and stopped. Send nothing yet.

> Tag this week's new sign-ups from the summer offer with summer-2026.

> How much did each course sell this month?

## Connect ChatGPT and Claude

1. [Get access to LearnWorlds on Cotlet.dev](https://cotlet.dev/signup?app=learnworlds).
2. Enable LearnWorlds in your Cotlet.dev console, open `https://account.learnworlds.com` in your Cloud Browser and sign in with your own school login. Complete any verification yourself.
3. Add `https://cotlet.dev/mcp/v1/connect` as a custom MCP connector in ChatGPT or Claude, then allow access.

## Connect Codex and Claude Code

Create an API key in the Cotlet.dev console and set `COTLET_API_KEY` in your environment.

```toml
[mcp_servers.cotlet]
url = "https://cotlet.dev/mcp/v1"
bearer_token_env_var = "COTLET_API_KEY"
```

```bash
claude mcp add --transport http cotlet https://cotlet.dev/mcp/v1/connect
```

Run `/mcp` in Claude Code and sign in to Cotlet.dev.

## Optional local MCP wrapper

Requires Python 3.10 or newer, standard library only. From this repository directory, run:

```bash
python learnworlds_agent.py
```

Configure your MCP client to start that command with `COTLET_API_KEY` in its environment. The wrapper provides seven browser helpers: `learnworlds_guide`, `learnworlds_status`, `learnworlds_open`, `learnworlds_read`, `learnworlds_click`, `learnworlds_type` and `learnworlds_close`. These are generic browser helpers; they do not call the LearnWorlds API. They use your signed-in Cloud Browser through Cotlet.dev.

## Questions

**Does LearnWorlds have an official MCP server?**

Yes. The live page reports an official MCP server in its initial beta, on Learning Center and higher plans, checked 20 November 2026. An admin with full rights turns it on in Settings, Developers, MCP.

**Is Cotlet.dev the same as LearnWorlds MCP?**

No. Cotlet.dev is an independent cloud browser service. Your agent uses the same LearnWorlds screens that your login can open.

**Which login should the agent use?**

An admin or team login with the permissions needed for the task. Payouts, billing and your LearnWorlds plan remain in your own hands.

## Access and control

You complete sign-in and any verification yourself. Disable LearnWorlds for agents, reset the browser or delete the key in the Cotlet console to revoke access. Cotlet.dev is independent and is not affiliated with LearnWorlds.