Md Gitlab Mcp — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Md Gitlab Mcp (Agent Skill) and scored it 45/100 (orange). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 1 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 flagged
A base64 string of 128+ characters appears in a documentation file. Encoded prompt injection hides the hostile instruction in base64 — invisible to keyword filters — and relies on the agent's ability to decode it at runtime. There is no normal authoring reason to embed a multi-hundred-byte base64 blob in skill docs.
*.sig, SIGNATURES) outside the documentation.Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
A fully-featured GitLab MCP Server tailored to behave like the GitHub MCP but specifically engineered for GitLab Data Centers, Instances, and Cloud.
md-gitlab-mcp is a Model Context Protocol (MCP) server written in TypeScript. It gives AI agents the powerful ability to interact deeply with GitLab via the REST API. This goes far beyond basic issue tracking and provides 25 advanced capabilities, identically structured to the famous GitHub MCP tools.
This server exports an extensive list of tools allowing AI agents to:
The default @modelcontextprotocol/server-gitlab provides very limited tools (around 9 commands) and completely lacks operations regarding code review, reading pull request/merge request threads, or resolving discussions.
md-gitlab-mcp bridges this gap, adopting GitHub's best-practice MCP schema standards and flawlessly translating semantics (projects -> repos, merge requests -> pull requests internally but honoring GitLab's API structure) while delivering complete access to GitLab's API.
Agents powered by this MCP will see properly annotated tools with parameter descriptions via zod-to-json-schema, meaning the AI inherently knows exactly how to query your repositories without hallucination.
You can directly add this to your mcp_settings.json:
{
"mcpServers": {
"gitlab": {
"command": "node",
"args": [
"/path/to/md-gitlab-mcp/build/index.js"
],
"env": {
"GITLAB_API_URL": "https://gitlab.example.dev/api/v4",
"GITLAB_PERSONAL_ACCESS_TOKEN": "your-access-token"
}
}
}
}This tool is handcrafted and maintained by Mikita Dusmikeev. Check out my work and connect on social media:
Created in 2026. Code like there's no tomorrow.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.