Skip to content
MCP server

Keystone

Keep a C4 architecture model of your codebase and hand your AI the slice it needs

Install
Pricing
Freemium
Platforms
Web, API
Runs
Remotely
Version
1.0.0
Listing updated
3 days ago

About Keystone

Keystone is a freemium MCP server for developer tools. It runs as a hosted remote server, so there is nothing to install locally.

Keystone keeps a model of your software as systems, services, components with public APIs, and databases down to tables and columns. Your AI coding assistant builds it by reading the repository in your own environment, so no source is uploaded; it sends Keystone elements, relationships, descriptions and evidence, which may contain code fragments.

Proposals arrive as a branch your team reviews, with field-by-field three-way merges. After code changes the AI can check the diagram for drift, by hand or from a CI job it generates for GitHub Actions or GitLab CI. The agent can request one module, what depends on an element, a path between two elements, or only what changed since last time.

The maker reports 68 to 77 percent fewer input tokens for a single-module question than sending the whole model, measured on one anonymized project. Scenarios, ADRs, antipattern review, live co-editing and exports to PNG, SVG, Mermaid, Structurizr DSL and Obsidian are included.

Install

Run in a terminal

claude mcp add --transport http keystone https://keystone-core.pro/mcp

Add to Settings → Connectors → Add custom connector

https://keystone-core.pro/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "keystone": {
      "url": "https://keystone-core.pro/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "keystone": {
      "type": "http",
      "url": "https://keystone-core.pro/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "keystone": {
      "serverUrl": "https://keystone-core.pro/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "keystone": {
      "type": "http",
      "url": "https://keystone-core.pro/mcp"
    }
  }
}

What it does

Model built from code
Your AI proposes elements from code, config and migrations.
Focused context
Returns a module, its API or a database instead of the whole graph.
Drift checks
Compare the diagram with code, manually or from CI.
Review by branch
Every AI proposal arrives as a branch you correct, merge or reject.
Decisions and scenarios
Attach ADRs to elements and play scenarios as sequence diagrams.
Open exports
PNG, SVG, Mermaid, Structurizr DSL and Obsidian notes.

Good for

  • Give an AI assistant architecture context for a task
  • Detect drift between diagrams and code
  • Document services and database tables from migrations
  • Review antipatterns with remediation plans
  • Share architecture decisions with a team

Pricing

Freemium

Free plan available; paid prices not published

Prices change. Check keystone-core.pro before you buy.

Keystone FAQ

Does Keystone get my source code?

You do not upload the repository; your AI reads it locally and sends architecture details, which can include code snippets.

Which tools work with it?

MCP clients including Claude Code, Cursor, VS Code with Copilot, Gemini CLI and Windsurf.

Is there a free plan?

The site says a free plan is available.

Can it run in my network?

The site says companies can also run Keystone inside their own network.

Is Keystone free?

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

How do I add Keystone to Claude Code?

Run this command in your terminal: claude mcp add --transport http keystone https://keystone-core.pro/mcp

What are the best alternatives to Keystone?

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

Alternatives to Keystone

  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