Skip to content
MCP server

Stele

Shared project memory for AI coding agents: decisions, lessons, risks and tasks as a graph

Install
Pricing
Freemium
Platforms
Web, API
Runs
Remotely
Version
1.0.0
Listing updated
1 day ago

About Stele

Stele is a freemium MCP server for knowledge and memory. It runs as a hosted remote server, so there is nothing to install locally.

Stele keeps a project's decisions, lessons, risks, documents, components and tasks in one knowledge graph that different agents can read and write. The hosted endpoint is the same URL for everyone; your account scopes it. Clients register themselves through OAuth 2.1, you approve once in a browser, and no key is pasted anywhere.

The point of connecting is reach: agents that never touch code, such as ChatGPT drafting a launch post or a phone assistant answering a customer, can read the same record as the agents doing the building. Tools let an agent recall what a project knows before acting, trace why a decision was made, capture new decisions and lessons, claim and complete tasks so two agents do not duplicate work, and hand off cleanly. Destructive operations sit in one tool that always asks first.

The free plan includes the hosted MCP, unlimited public projects, one private project, 500 knowledge writes a month and unlimited reads. The connector plugin is MIT licensed.

Install

Run in a terminal

claude mcp add --transport http stele https://app.stele-ai.dev/api/mcp

Add to Settings → Connectors → Add custom connector

https://app.stele-ai.dev/api/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "stele": {
      "url": "https://app.stele-ai.dev/api/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "stele": {
      "type": "http",
      "url": "https://app.stele-ai.dev/api/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "stele": {
      "serverUrl": "https://app.stele-ai.dev/api/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "stele": {
      "type": "http",
      "url": "https://app.stele-ai.dev/api/mcp"
    }
  }
}

What it does

Project recall
Agents check what the project knows before acting.
Decision tracing
Follow why something was decided and what it replaced.
Task coordination
Create, claim and complete tasks across agents.
Safe tool tiers
Read, write and destructive tools are kept separate.
Browser sign-in
OAuth with no tokens to paste.
ChatGPT cards
ChatGPT can show tasks and entries as cards.

Good for

  • Give ChatGPT context from your coding project
  • Record decisions so later agents do not repeat debates
  • File a task from your phone
  • Hand over work between agent sessions

Pricing

Freemium

Free plan with 500 writes a month; paid Indie, Pro and Team tiers (prices not shown)

Prices change. Check stele-ai.dev before you buy.

Stele FAQ

Is it free?

Yes, the hosted MCP is included on the free plan with no card; paid plans add private projects and limits.

Are reads metered?

No. Search and retrieval are unlimited on every plan.

What does a knowledge write count?

Adding something new; editing or archiving does not count.

Do I install anything?

No, just paste the endpoint URL and sign in.

Is Stele free?

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

How do I add Stele to Claude Code?

Run this command in your terminal: claude mcp add --transport http stele https://app.stele-ai.dev/api/mcp

What are the best alternatives to Stele?

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

Alternatives to Stele

  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