BuildBook MCP server
# BuildBook MCP server
Connect BuildBook to ChatGPT, Claude and other AI agents through Cotlet.dev. Work with estimates, schedules, daily logs, bills and invoices in your signed-in BuildBook account. Review write operations before approving them.
## What you can ask
> List overdue tasks across my active remodeling projects.
> Draft an estimate for the Maple Street bath remodel and show me the lines before saving.
> Read the project budget report and summarize recorded costs.
## Connect in 2 minutes
Get your API key at https://cotlet.dev/app and export it as `COTLET_API_KEY`.
### ChatGPT
Settings → Apps → add connector URL `https://cotlet.dev/mcp/v1/connect`, then sign in with Cotlet.dev. No API key needed.
### Claude Code
```bash
claude mcp add --transport http cotlet https://cotlet.dev/mcp/v1 --header "Authorization: Bearer $COTLET_API_KEY"
```
### 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 BuildBook 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.
Run the included Python module with Python 3.10 or later:
```bash
python3 buildbook_agent.py serve
```
Or with Docker:
```bash
docker build -t buildbook-mcp .
docker run -i --rm -e COTLET_API_KEY buildbook-mcp
```
Tools: `buildbook_guide`, `buildbook_status`, `buildbook_open`, `buildbook_read`, `buildbook_click`, `buildbook_type`, `buildbook_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+, with no third-party runtime dependencies. Clone this repository and run:
```bash
export COTLET_API_KEY="your-key"
python3 buildbook_agent.py connect # ready config for Claude Code, Cursor, ChatGPT
python3 buildbook_agent.py status # is BuildBook ready in your Cotlet.dev account
python3 buildbook_agent.py guide # BuildBook screens and sign-in link
python3 buildbook_agent.py read # open BuildBook, read the page, close the browser
python3 buildbook_agent.py tools # list Cotlet.dev tools
python3 buildbook_agent.py serve # start the stdio MCP server (same as no argument)
```
## Questions from BuildBook users
**How do I connect my account?**
Request access at https://cotlet.dev/signup?app=buildbook, sign in to BuildBook in your Cotlet.dev Cloud Browser and enable it for agents.
**Can I review changes?**
The hosted integration asks for confirmation before write calls by default. Saving an estimate as a draft and sharing it with the client are separate actions.
**Does BuildBook provide an API?**
The live Cotlet.dev integration page reports no public BuildBook API or verified official MCP server as checked on 13 November 2026. BuildBook documents its own QuickBooks Online integration.
**Is Cotlet.dev affiliated with BuildBook?**
No. Cotlet.dev is independent and not affiliated with BuildBook.
## Learn more
- [buildbook integration and hosted tools](https://cotlet.dev/integrations/buildbook-api-mcp-server)
---
More: https://cotlet.dev/integrations/buildbook-api-mcp-server?utm_source=github&utm_medium=readme&utm_campaign=buildbook
Request access: https://cotlet.dev/signup?app=buildbook&utm_source=github&utm_medium=readme&utm_campaign=buildbook
## Workflows on the hosted integration
Examples documented on the live integration page:
- `view_pro_dashboard`
- `view_project_activity`
- `view_client_dashboard`
- `search_projects`
- `read_project_chat`
- `list_people`
- `manage_opportunities`
- `manage_estimates`
- `configure_estimate_defaults`
- `manage_price_book`
- `manage_tax_rates`
- `manage_projects`
- `manage_project_templates`
- `manage_schedule`
- `view_schedule`
- `manage_tasks`
- `manage_task_statuses`
- `print_tasks`
- `manage_daily_logs`
- `manage_project_files`
- `invite_people`
- `view_project_financials`
- `manage_budget`
- `manage_bills_and_expenses`
- `export_budget_report`
- `manage_vendors`
- `manage_change_orders`
- `manage_selections`
- `manage_invoices`
- `manage_portfolio`
The local stdio adapter exposes seven browser helpers listed above. These helpers forward to Cotlet.dev; the workflow operations are documented on the hosted integration page.
TDQS
Scored across 7 tools
Each tool maps to a distinct phase of a browser-automation workflow: guide (instructions), status (account/connection check), open (launch session), read (inspect page), click, type, and close. There is no meaningful overlap between any two tools, and the descriptions reinforce clear boundaries.
All tools share a uniform buildbook_ prefix followed by a single lowercase verb or noun (guide, status, open, read, click, type, close). The pattern is predictable and readable throughout.
Seven tools is well-scoped for a browser-automation server: a guide, a health check, and the five core session primitives (open/read/click/type/close). Nothing feels redundant or missing at the count level.
The surface covers the full basic browser lifecycle: open a page, read its contents, click, type, and close the session. Minor gaps exist—no explicit navigation to an arbitrary URL, scrolling, or dropdown/select handling—but agents can likely work around these via click/type and the guide.