Skip to main content
Glama
README.md
# LMN MCP server

Connect LMN to ChatGPT, Claude and other AI agents through Cotlet.dev. Review landscaping estimates, jobs, schedules, timesheets and billing in your signed-in LMN account. Hosted write tools ask for confirmation by default.

## What you can ask

> List estimates waiting for proposal follow-up.

> Review yesterday's timesheets and show me alerts before approval.

> Preview invoices ready to generate for the Maintenance group this month.

## 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 LMN 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 lmn_agent.py serve
```

Or with Docker:

```bash
docker build -t lmn-mcp .
docker run -i --rm -e COTLET_API_KEY lmn-mcp
```

Tools: `lmn_guide`, `lmn_status`, `lmn_open`, `lmn_read`, `lmn_click`, `lmn_type`, `lmn_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 lmn_agent.py connect   # ready config for Claude Code, Cursor, ChatGPT
python3 lmn_agent.py status    # is LMN ready in your Cotlet.dev account
python3 lmn_agent.py guide     # LMN screens and sign-in link
python3 lmn_agent.py read      # open LMN, read the page, close the browser
python3 lmn_agent.py tools     # list Cotlet.dev tools
python3 lmn_agent.py serve     # start the stdio MCP server (same as no argument)
```

## Questions from LMN users

**How do I connect my account?**

Request access at https://cotlet.dev/signup?app=lmn, sign in to LMN in your Cotlet.dev Cloud Browser and enable it for agents.

**Can I review changes first?**

Hosted write tools ask for confirmation by default. Keep confirmation enabled for payment runs and saved-method payments, which move money.

**Does this use the LMN Zapier API key?**

No. Cotlet.dev works through your signed-in Cloud Browser. The live integration page compares the documented LMN Zapier integration with the hosted workflows.

**Is Cotlet.dev affiliated with LMN?**

No. Cotlet.dev is independent and not affiliated with LMN or Granum.

## Learn more

- [LMN integration and hosted tools](https://cotlet.dev/integrations/lmn-api-mcp-server)

---

More: https://cotlet.dev/integrations/lmn-api-mcp-server?utm_source=github&utm_medium=readme&utm_campaign=lmn

Request access: https://cotlet.dev/signup?app=lmn&utm_source=github&utm_medium=readme&utm_campaign=lmn

## Workflows on the hosted integration

Examples documented on the live integration page:

- `manage_budget`
- `manage_price_list`
- `manage_services`
- `list_estimates`
- `manage_estimates`
- `send_proposal`
- `configure_lead_form`
- `manage_jobs`
- `manage_change_orders`
- `manage_schedule`
- `stop_service`
- `manage_work_calendars`
- `manage_timesheets`
- `manage_payroll_settings`
- `manage_sms`
- `generate_invoices`
- `manage_invoices`
- `manage_invoice_groups`
- `configure_invoice_automation`
- `list_payments`
- `record_payment`
- `run_payments`
- `request_payment_method`

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

A3.7/5.0

Scored across 7 tools

Disambiguation5/5

Each tool targets a distinct phase of browser work: guide (onboarding docs), status (readiness check), open (session start), read (page inspection), click and type (element interaction), and close (session teardown). The lmn_guide vs lmn_status boundary is slightly close but descriptions clearly separate documentation from account/connection checks.

Naming Consistency5/5

All seven tools use the same lmn_ prefix followed by a concise snake_case verb or state name (open, read, click, type, close, guide, status). The pattern is predictable and consistent throughout.

Tool Count5/5

Seven tools is well-scoped for a browser automation surface; each tool corresponds to a discrete step (start, inspect, interact, finish) with no redundant primitives. Nothing feels padded or missing from the core interaction loop.

Completeness4/5

The set covers the full session lifecycle—status, open, read, click, type, close—plus a guide, which is strong for the stated purpose. Minor gaps remain: no explicit navigation/goto for arbitrary URLs, no scroll or wait primitive, and no select/dropdown helper, which could matter for multi-page LMN workflows.

Maintenance

ActivityMaintained
ResponsivenessOutdated data