# Zentrik

A remote MCP server that lets an agent read the product decisions, customer evidence and workspace context your team keeps in Zentrik. Useful when plans and edits should rest on data you already collected.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI
- Author: Zentrik-AI
- License: Apache-2.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/Zentrik-AI/agent-plugins
- Updated: 2026-10-21
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/zentrik

## What it does

The endpoint is Streamable HTTP at `https://zentrik.ai/mcp`. Sign-in goes through Zentrik OAuth, and each connection is locked to one workspace you pick, inside the role you hold and the scopes you granted. No credentials and no local executables ship with the plugin; everything travels over the network.

Four workflows come bundled. `brief-product-work` feeds existing evidence into planning and review, `set-up-product-workspace` applies a bounded setup change you approved first, `import-product-evidence` files calls, tickets, reviews or research behind a confirmation step, and `manage-product-study` drafts findings for a private Study. Writes such as importing evidence stop at your Zentrik role and scopes.

You need a Zentrik account and membership in at least one workspace. An API key is not part of the setup. Without a workspace there is nothing to bind the session to, so pick one right after the browser sign-in. VS Code users paste a JSON block into `.vscode/mcp.json`; Codex goes through the adapter at `com.openai/zentrik`; server-to-server jobs belong to the REST API under `/external/v1`, which is not bundled here.

## Requirements
- Remote server: `https://zentrik.ai/mcp`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http mcp https://zentrik.ai/mcp
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add zentrik --url https://zentrik.ai/mcp
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "zentrik": {
            "url": "https://zentrik.ai/mcp"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "zentrik": {
            "type": "http",
            "url": "https://zentrik.ai/mcp"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http zentrik https://zentrik.ai/mcp
```

## When to use it
- ground a task plan in the customer evidence already stored in Zentrik
- check which initiative a branch belongs to before editing it
- import reviews and tickets as evidence with approval before any write
- apply a small, approved setup change to a workspace
- draft Study findings from collected evidence

## Example prompts
- Pull the customer quotes about bulk export and cite where each one came from.
- Before I edit this branch, sum up what the related initiative was meant to do.
- File these support tickets as evidence, but show me the writes first.

## Similar
- [Fathom MCP Server](https://affpapa.org/en/nevibecode/mcp/fathom-mcp-server.md) — MCP server · ★ 16 repo stars
- [Pairoa](https://affpapa.org/en/nevibecode/mcp/pairoa.md) — MCP server · ★ 3 repo stars
- [Everway MCP Server](https://affpapa.org/en/nevibecode/mcp/everway-mcp-server.md) — MCP server · ★ 1 repo stars
- [VerifEye](https://affpapa.org/en/nevibecode/mcp/verifeye.md) — MCP server · ★ 1 repo stars
- [FlightQueue](https://affpapa.org/en/nevibecode/mcp/flightqueue.md) — MCP server
- [mcpoauthscopegate-mcp](https://affpapa.org/en/nevibecode/mcp/mcpoauthscopegate-mcp.md) — MCP server

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
