FRENES

Developer tools · MCP server

Primitiv: MCP server, setup and security

Primitiv is an MCP server published on GitHub that lets an AI agent reconcile design rules across Figma, code, Storybook and token files. Its repository has 21 stars and its last commit is from October 21, 2026, but it declares no licence.

Actively maintainedNo licence

What Primitiv does

According to its repository: “Local-first design contract layer: scans Figma, code, Storybook and token files, reconciles conflicts and serves a single source of truth for design rules.”

An MCP server gives an AI agent extra tools it calls during the conversation: the agent reads their description, decides to use them and gets the result back. Primitiv is listed under Developer tools, a category of 475 servers in the directory, where it ranks #332 by stars.

Topics declared by the repository: ai-agents, design-system, design-tokens, figma, llm-tools, llm-tools-directory-tree.

Installing and connecting Primitiv

No package or remote server is declared in the official MCP registry: you install it from source, following the repository README.

Like any MCP server, you declare it in your client's configuration (Claude Desktop, Claude Code, Cursor, VS Code, Codex…): a name, the command that starts it or its URL, and any environment variables. The client then starts it and lists its tools. Project language: TypeScript. Listed systems: macOS, Windows, Linux.

Security: what to check

An MCP server runs code with the rights you give it, and the model reads its tool descriptions: a malicious or compromised server can hijack the agent. Read the code, or pick a maintained and widely used server, before connecting it.

Points found for Primitiv:

  • No licence: without a licence the code is not legally reusable, even if it is public. Avoid it in a company setting without the author's agreement.

Good practice: grant only the rights needed (least privilege), require human approval for irreversible actions, and run the AI agent security check.

Project activity and maintenance

Created on March 14, 2026, the AI-by-design/primitiv repository has 21 stars and 1 forks, with 0 open issues. It has more stars than 39% of the servers in the directory. It received changes this month.

Stars measure attention, not quality: a recent, heavily promoted project can collect them fast, while a useful but niche server will have few.

StorybookStorybook : 91.4K91.4KFront End ChecklistFront End Checklist : 74.6K74.6KTldrawTldraw : 51.1K51.1KCodebase MemoryCodebase Memory : 47.4K47.4KCCXTCCXT : 44.5K44.5KPostHog MCP ServerPostHog MCP Server : 40.4K40.4KClaude task masterClaude task master : 28.3K28.3KAgent Skills Search ServerAgent Skills Search Server : 26.6K26.6KPrimitivPrimitiv : 2121
GitHub stars of the most followed servers in the category (log scale); Primitiv in orange. Checked October 25, 2026.

Verdict

Primitiv is still a modest project: read its code and compare it with the alternatives before using it.

Strengths

  • Active project: recent commit

Limitations

  • No licence declared
  • Install from source

Avoid if…

  • Company use without the author's agreement

Frequently asked questions

How do I install Primitiv?

No package is declared in the MCP registry: clone the AI-by-design/primitiv repository, follow its README, then declare the start command in your client's MCP configuration.

Is Primitiv free?

The code is public but has no declared licence: it is not formally free to reuse.

Is Primitiv safe?

Points to watch: no licence. Give it minimum rights and approve sensitive actions.

Is it an official server?

Nothing says so: it is a project published by AI-by-design on GitHub. Check who maintains it before giving it access.

What are the alternatives to Primitiv?

In the same category: Radar Kubernetes MCP Server, Repowise, Jscpd, and Apify MCP server. Radar Kubernetes MCP Server and Repowise score higher on our grid.

Related articles

Neutrality. No affiliate links. No vendor pays to be reviewed, rated or ranked. Reviews are directed by François Chollet, neutral and factual, and every data point is dated and sourced. How we score