Skip to main content
Glama

Gripp MCP

Model Context Protocol server for the Gripp API 3.0 endpoint at https://api.gripp.com/public/api3.php.

The Gripp API is JSON-RPC-like: requests are POSTed as a batch array, and the API token is sent as Authorization: Bearer <token>.

Setup

npm install
npm run fetch:metadata
npm run build

Set your Gripp API token before starting the server:

export GRIPP_API_TOKEN="your-token"
npm start

Optional environment variables:

GRIPP_API_URL=https://api.gripp.com/public/api3.php
GRIPP_REQUEST_TIMEOUT_MS=30000

GRIPP_API_URL is intentionally restricted to the https://api.gripp.com/ host.

Related MCP server: mcp-github

MCP Client Config

Example stdio config:

{
  "mcpServers": {
    "gripp": {
      "command": "node",
      "args": ["/absolute/path/to/MCP Gripp/dist/src/cli.js"],
      "env": {
        "GRIPP_API_TOKEN": "your-token"
      }
    }
  }
}

After publishing or linking the package, you can also run the gripp-mcp binary.

Vercel Remote MCP

This repo also exposes a Streamable HTTP MCP endpoint for Vercel through a Next.js route:

https://your-vercel-domain.vercel.app/api/mcp

For Claude custom connectors, use the /api/mcp URL. The root URL only returns a health response.

Set this environment variable in Vercel before using Gripp tools:

GRIPP_API_TOKEN=your-token

The remote endpoint can also read a Gripp token from Authorization: Bearer <token> or x-gripp-api-token, but Claude's custom connector UI does not currently provide a simple custom-header field. For Claude, the practical setup is to store GRIPP_API_TOKEN in the Vercel project environment.

Tools

  • gripp_list_entities: list available Gripp entities and method names.

  • gripp_describe_entity: inspect fields, enum values, references, methods, and examples for one entity.

  • gripp_get: call <entity>.get with filters and options.

  • gripp_getone: call <entity>.getone with filters.

  • gripp_create: call <entity>.create; requires confirm: true.

  • gripp_update: call <entity>.update; requires confirm: true.

  • gripp_delete: call <entity>.delete; requires confirm: true.

  • gripp_call: call any documented Gripp method by full name.

  • gripp_batch: execute multiple documented Gripp calls in one transaction.

Non-read methods are blocked unless confirm: true is provided. This also applies to gripp_call and each item in gripp_batch.

Examples

Find companies by name:

{
  "entity": "company",
  "filters": [
    {
      "field": "company.companyname",
      "operator": "like",
      "value": "%Acme%"
    }
  ],
  "options": {
    "paging": {
      "firstresult": 0,
      "maxresults": 10
    },
    "orderings": [
      {
        "field": "company.companyname",
        "direction": "asc"
      }
    ]
  }
}

Call a nonstandard read method:

{
  "method": "company.getCompanyByCOC",
  "params": ["12345678"]
}

Create a tag:

{
  "entity": "tag",
  "fields": {
    "name": "Imported"
  },
  "confirm": true
}

Metadata

The committed metadata snapshot is generated from the public API docs page:

npm run fetch:metadata

Refresh it when Gripp updates the API docs.

Install Server
F
license - not found
A
quality
C
maintenance

Maintenance

Maintainers
Response time
Release cycle
Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/maxancedecot/GrippMCP'

If you have feedback or need assistance with the MCP directory API, please join our Discord server