Skip to content
MCP server

Systemdox

Feed an agent your architecture decisions, specs and checks, and record new ones

Runs locallyStatus and history
Install
Pricing
Freemium
Platforms
Web, CLI, API
Runs
Locally
Version
0.1.66
Listing updated
3 days ago

About Systemdox

Systemdox is a freemium MCP server for developer tools. It runs locally on your machine and is started by your AI client.

SystemDox is a documentation platform built around the idea that architecture knowledge should be in a coding agent's context at the moment code is written. Its MCP server connects Claude Code, Codex, Cursor or any MCP client to a workspace holding standards, specs, architecture decision records, checks and project rules versioned in git.

get_context ranks the checks, specs and docs relevant to the task and fetches their bodies within a token budget. In the other direction, report_decision and update_spec let an agent record decisions and spec changes as it works, so decision records appear as a by-product of shipping. The API key is the boundary: a department-scoped key sees only that space's material.

Connect by signing in with a passkey (read-only unless you allow writes), through a per-workspace hosted endpoint with a Bearer key, or locally via the npm package. SystemDox is in private preview and runs the vendor's own engineering today, so a workspace has to be requested.

Install

Run in a terminal

claude mcp add systemdox --env SYSTEMDOX_API_KEY='<SYSTEMDOX_API_KEY>' -- npx -y @systemdox/mcp-server

Add to claude_desktop_config.json

{
  "mcpServers": {
    "systemdox": {
      "command": "npx",
      "args": [
        "-y",
        "@systemdox/mcp-server"
      ],
      "env": {
        "SYSTEMDOX_API_KEY": "<SYSTEMDOX_API_KEY>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "systemdox": {
      "command": "npx",
      "args": [
        "-y",
        "@systemdox/mcp-server"
      ],
      "env": {
        "SYSTEMDOX_API_KEY": "<SYSTEMDOX_API_KEY>"
      }
    }
  }
}

Add to .vscode/mcp.json

{
  "servers": {
    "systemdox": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@systemdox/mcp-server"
      ],
      "env": {
        "SYSTEMDOX_API_KEY": "<SYSTEMDOX_API_KEY>"
      }
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "systemdox": {
      "command": "npx",
      "args": [
        "-y",
        "@systemdox/mcp-server"
      ],
      "env": {
        "SYSTEMDOX_API_KEY": "<SYSTEMDOX_API_KEY>"
      }
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "systemdox": {
      "command": "npx",
      "args": [
        "-y",
        "@systemdox/mcp-server"
      ],
      "env": {
        "SYSTEMDOX_API_KEY": "<SYSTEMDOX_API_KEY>"
      }
    }
  }
}

Configuration

  • SYSTEMDOX_API_KEYSystemDox API key (generate in workspace settings). Scopes every tool call to the key's tenant.Required
  • SYSTEMDOX_API_URLREST API base URL override (optional).

Tools this server provides (3)

Taken from the project documentation; the running server may differ.

get_context
Ranks and fetches checks, specs and docs relevant to a task within a token budget.
report_decision
Records an architecture decision from an agent.
update_spec
Updates a spec from an agent.

What it does

Write-time context
get_context returns ranked checks, specs and docs within a token budget.
Decisions written back
report_decision and update_spec let agents record decisions and spec updates.
Key-scoped access
A key limits tools to its tenant or department space.
Sign-in connection
Workspace owners can connect with a passkey instead of a key.
Hosted or local
Per-workspace HTTP endpoint or the @systemdox/mcp-server npm package.
Capture and plan
The platform also generates documents from voice notes and drafts plans for a GitHub Project board.

Good for

  • Give every coding agent the same architecture rules
  • Record an architecture decision while a feature is built
  • Keep a spec in sync with the code changes an agent makes
  • Publish engineering docs as a browsable knowledge base

Pricing

Freemium

Free for one repository; Business plan by quote; private preview

Prices change. Check systemdox.com before you buy.

Systemdox FAQ

Is it open to everyone?

No. The site says it is in private preview and plans open at public launch.

What does the Free plan include?

One connected Git repository, included AI usage, a published knowledge base and MCP access.

How do I authenticate?

With a SystemDox API key, or by signing in with a passkey as a workspace owner or admin.

Should I commit my key?

No. The README tells you to reference it through an environment variable.

Is Systemdox free?

Systemdox has a free plan, with paid plans for more usage or features.

How do I add Systemdox to Claude Code?

Run this command in your terminal: claude mcp add systemdox --env SYSTEMDOX_API_KEY='<SYSTEMDOX_API_KEY>' -- npx -y @systemdox/mcp-server

What are the best alternatives to Systemdox?

The closest alternatives to Systemdox are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to Systemdox

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools