GitLab MCP server for AI agents
Let Claude, ChatGPT or your own agent read GitLab: 4 typed functions through one MCP server. No GitLab API key needed.
- Site
- gitlab.com
- Functions
- 4
- Sign-in
- Not needed
- Works with
- MCP, HTTP, Node, Python
Connect
An MCP server for GitLab alone, with one tool per function. Paste it into Claude, ChatGPT, Cursor or any MCP client as a remote server.
https://api.bowmark.ai/mcp/site/gitlabOne MCP server for GitLab and every other site Bowmark covers. Your agent looks up what it needs and calls it.
https://api.bowmark.ai/mcpSet it up in your clientBoth need a Bowmark API key. The quickstart gets you one and a first result in a few minutes.
Functions
Open a function to see its arguments, what it returns and how to call it.
getProjectGets one public GitLab project by numeric id or path, such as gitlab-org/gitlab.Arguments
Name Type Required projectstringYes Returns
GitlabProjectinterface GitlabProject { id: number; pathWithNamespace: string; name: string; description: string | null; webUrl: string; stars: number; forks: number; defaultBranch: string | null; lastActivityAt: string | null }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/gitlab/getProjectCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.gitlab.getProject("gitlab-org/gitlab");import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.gitlab.getProject("gitlab-org/gitlab"))curl -X POST https://api.bowmark.ai/v1/sites/gitlab/getProject \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"project":"gitlab-org/gitlab"}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.gitlab.getProject("gitlab-org/gitlab");searchIssuesSearches issues in one public GitLab project by project path or id, with an optional text query.Arguments
Name Type Required projectstringYes querystringNo limitnumberNo Returns
GitlabWorkItem[]interface GitlabWorkItem { iid: number; projectId: number; title: string; description: string | null; state: string; author: string | null; webUrl: string; createdAt: string | null; updatedAt: string | null }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/gitlab/searchIssuesCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.gitlab.searchIssues({ project: "gitlab-org/gitlab", query: "bug", limit: 2 });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.gitlab.searchIssues({"project": "gitlab-org/gitlab", "query": "bug", "limit": 2}))curl -X POST https://api.bowmark.ai/v1/sites/gitlab/searchIssues \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"project":"gitlab-org/gitlab","query":"bug","limit":2}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.gitlab.searchIssues({ project: "gitlab-org/gitlab", query: "bug", limit: 2 });searchMergeRequestsSearches merge requests in one public GitLab project by project path or id, with an optional text query.Arguments
Name Type Required projectstringYes querystringNo limitnumberNo Returns
GitlabWorkItem[]interface GitlabWorkItem { iid: number; projectId: number; title: string; description: string | null; state: string; author: string | null; webUrl: string; createdAt: string | null; updatedAt: string | null }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/gitlab/searchMergeRequestsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.gitlab.searchMergeRequests({ project: "gitlab-org/gitlab", query: "bug", limit: 2 });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.gitlab.searchMergeRequests({"project": "gitlab-org/gitlab", "query": "bug", "limit": 2}))curl -X POST https://api.bowmark.ai/v1/sites/gitlab/searchMergeRequests \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"project":"gitlab-org/gitlab","query":"bug","limit":2}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.gitlab.searchMergeRequests({ project: "gitlab-org/gitlab", query: "bug", limit: 2 });searchProjectsFinds public GitLab projects by name or description, returning a project path for follow-up lookups.Arguments
Name Type Required querystringYes limitnumberNo Returns
GitlabProject[]interface GitlabProject { id: number; pathWithNamespace: string; name: string; description: string | null; webUrl: string; stars: number; forks: number; defaultBranch: string | null; lastActivityAt: string | null }HTTP endpoint
POST https://api.bowmark.ai/v1/sites/gitlab/searchProjectsCall it
import { bowmark } from "@bowmark/web"; const result = await bowmark.providers.gitlab.searchProjects({ query: "gitlab", limit: 2 });import asyncio from bowmark_web import bowmark result = asyncio.run(bowmark.providers.gitlab.searchProjects({"query": "gitlab", "limit": 2}))curl -X POST https://api.bowmark.ai/v1/sites/gitlab/searchProjects \ -H "Authorization: Bearer $BOWMARK_API_KEY" \ -H "Content-Type: application/json" \ -d '{"args":{"query":"gitlab","limit":2}}'// The body of a run: the MCP run tool, or POST /v1/run return await bowmark.providers.gitlab.searchProjects({ query: "gitlab", limit: 2 });
Use it from code
- Node
npm i @bowmark/web - Python
pip install bowmark-web - HTTP
POST /v1/sites/gitlab/<function> - OpenAPI
/v1/sites/gitlab/openapi.json