Studio Ninja MCP
# Studio Ninja MCP server for Claude and ChatGPT
Studio Ninja has no public API or Zapier app. Cotlet runs a hosted Studio Ninja MCP server with 9 tools for leads, jobs and invoices, in ChatGPT and Claude.
## What you can ask
> Log a new enquiry as a lead with its job type and date.
> List leads that have gone quiet and prepare follow-ups.
> Book a lead as a job with its Studio Ninja workflow.
> Send the contract or invoice after I approve it.
> List overdue invoices and payments due soon.
## Connect in 2 minutes
Get your API key at https://cotlet.dev/app and export it as `COTLET_API_KEY`.
### Claude Code
```bash
claude mcp add --transport http cotlet https://cotlet.dev/mcp/v1 --header "Authorization: Bearer $COTLET_API_KEY"
```
### ChatGPT
Settings → Apps → add connector URL `https://cotlet.dev/mcp/v1`, then sign in with Cotlet.dev. No API key needed.
### Cursor
Copy `.cursor/mcp.json` from this repo into your project:
```json
{
"mcpServers": {
"cotlet": {
"url": "https://cotlet.dev/mcp/v1",
"headers": {
"Authorization": "Bearer ${COTLET_API_KEY}"
}
}
}
}
```
## Run it locally (stdio)
This repo is a real MCP server. It speaks MCP over stdio, lists its Studio Ninja tools without any key, and forwards every tool call to the hosted Cotlet.dev server. The hosted URL above stays the recommended way. Use this when your client only starts local servers.
```json
{"mcpServers":{"studio-ninja":{"command":"uvx","args":["studio-ninja-mcp"],"env":{"COTLET_API_KEY":"your-key"}}}}
```
Or with Docker:
```bash
docker build -t studio-ninja-mcp .
docker run -i --rm -e COTLET_API_KEY studio-ninja-mcp
```
Tools: `studio_ninja_guide`, `studio_ninja_status`, `studio_ninja_open`, `studio_ninja_read`, `studio_ninja_click`, `studio_ninja_type`, `studio_ninja_close`. Without `COTLET_API_KEY` a tool call answers: "Set COTLET_API_KEY. Get a key at https://cotlet.dev/app".
## Command line
Python 3.10+, nothing else to install. From PyPI:
```bash
pip install studio-ninja-mcp # or run it once: uvx studio-ninja-mcp connect
studio-ninja-mcp connect
```
Or straight from this repo:
```bash
export COTLET_API_KEY="your-key"
python3 studio_ninja_agent.py connect # ready config for Claude Code, Cursor, ChatGPT
python3 studio_ninja_agent.py status # is Studio Ninja ready in your Cotlet.dev account
python3 studio_ninja_agent.py guide # Studio Ninja screens and sign-in link
python3 studio_ninja_agent.py read # open Studio Ninja, read the page, close the browser
python3 studio_ninja_agent.py tools # list Cotlet.dev tools
python3 studio_ninja_agent.py serve # start the stdio MCP server (same as no argument)
```
## Questions from Studio Ninja users
**Does Studio Ninja have an official MCP server or API?**
No. Studio Ninja has no official MCP server, public API or Zapier app (checked 8 November 2026). Cotlet is an independent hosted connection.
**How do I connect Studio Ninja to ChatGPT or Claude?**
Sign up at cotlet.dev, sign in to Studio Ninja once in its Cloud Browser and add Cotlet to ChatGPT or Claude, following the steps in your console.
**What can my AI agent do in Studio Ninja?**
It can log enquiries, follow up quiet leads, book jobs with their workflow, send contracts, send invoices and list overdue payments. Anything that reaches another person waits for your approval.
**Can it send an invoice or contract?**
Yes. Cotlet prepares and sends the Studio Ninja invoice or contract only after you approve the action.
**Does it work with Studio Ninja workflows?**
Yes. Booking a job applies the job type workflow as usual, and your agent can complete workflow tasks.
**Is Cotlet affiliated with Studio Ninja?**
No. Cotlet is independent and not affiliated with Studio Ninja or its owner Captura.
## Find it on
- [Studio Ninja on Cotlet.dev](https://cotlet.dev/integrations/studio-ninja-api-mcp-server?utm_source=github&utm_medium=readme&utm_campaign=studio-ninja&utm_content=find)
- [PyPI package studio-ninja-mcp](https://pypi.org/project/studio-ninja-mcp/)
- [Odel](https://odel.app/modules/cotlet/studio-ninja)
- [mcpbeat](https://mcpbeat.com/mcp-servers/cotlet/studio-ninja/)
- [Step-by-step guide (PDF) on Internet Archive](https://archive.org/details/cotlet-connect-studio-ninja-to-ai-agents-guide)
## Also works with
- [All apps connected through Cotlet.dev](https://github.com/cotlet-connect)
---
More: https://cotlet.dev/integrations/studio-ninja-api-mcp-server?utm_source=github&utm_medium=readme&utm_campaign=studio-ninja
Start free: https://cotlet.dev/signup?app=studio-ninja&utm_source=github&utm_medium=readme&utm_campaign=studio-ninja
TDQS
Scored across 7 tools
Each tool maps to a distinct step in a browser session lifecycle: readiness check (status), documentation (guide), navigation (open), inspection (read), and interaction (click/type/close). The closest pair, status vs. guide, is clearly separated by their descriptions (technical account/connection check vs. product how-to). No two tools could plausibly be selected for the same action.
All seven tools share the studio_ninja_ snake_case prefix with a single-word suffix, giving a highly predictable pattern. Minor deviation: status and guide are nouns while the rest are verbs (open/read/click/type/close), though the noun forms are conventional for informational tools.
Seven tools is well-scoped for a browser-driven SaaS automation surface: one per lifecycle stage (status, guide, open, read, click, type, close) with no redundancy. Nothing feels padded or thin.
Session lifecycle, navigation, page reading, and element interaction are all covered, forming a coherent read-then-act loop. Minor gaps remain: no explicit scroll or wait-for-load primitive for long/dynamic pages (e.g. overdue payment lists), no back/history navigation, and no dedicated select/keyboard action beyond click and type.