Skip to content
MCP server

scree

Local semantic and keyword search over the markdown in a repo or notes folder

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, Linux, macOS
Runs
Locally
Version
0.4.1
Listing updated
2 days ago

About scree

scree is an open-source MCP server for knowledge and memory. It runs locally on your machine and is started by your AI client.

scree is a single static Go binary that gives Claude Code search over READMEs, design docs, ADRs, agent instructions and notes, though not source code. It splits each markdown file by heading, embeds the pieces through OpenRouter with zero data retention requested, and keeps chunks, vectors and a keyword index in a SQLite file on your machine.

The agent calls search_docs and gets hits with a path, heading trail and line range, so it reads only those lines instead of whole files. Matching is by meaning and, for identifiers such as PURGE_OLD_RECORDS or PageFetcher, by exact words too. Without an API key it still indexes for keyword search.

The site gives rough running costs: about $0.0025 per MB of markdown to index once, only changed files re-embedded, and 13 to 20 MB of idle memory per process. A repo is only indexed if its own .mcp.json names scree mcp, and the README warns that approving a project server means trusting that repo. It is MIT licensed.

Install

Run in a terminal

claude mcp add scree --env OPENROUTER_API_KEY='<OPENROUTER_API_KEY>' -- docker run -i --rm -e OPENROUTER_API_KEY ghcr.io/jrapakko/scree:0.4.1

Add to claude_desktop_config.json

{
  "mcpServers": {
    "scree": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "OPENROUTER_API_KEY",
        "ghcr.io/jrapakko/scree:0.4.1"
      ],
      "env": {
        "OPENROUTER_API_KEY": "<OPENROUTER_API_KEY>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "scree": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "OPENROUTER_API_KEY",
        "ghcr.io/jrapakko/scree:0.4.1"
      ],
      "env": {
        "OPENROUTER_API_KEY": "<OPENROUTER_API_KEY>"
      }
    }
  }
}

Add to .vscode/mcp.json

{
  "servers": {
    "scree": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "OPENROUTER_API_KEY",
        "ghcr.io/jrapakko/scree:0.4.1"
      ],
      "env": {
        "OPENROUTER_API_KEY": "<OPENROUTER_API_KEY>"
      }
    }
  }
}

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

{
  "mcpServers": {
    "scree": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "OPENROUTER_API_KEY",
        "ghcr.io/jrapakko/scree:0.4.1"
      ],
      "env": {
        "OPENROUTER_API_KEY": "<OPENROUTER_API_KEY>"
      }
    }
  }
}

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

{
  "mcpServers": {
    "scree": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "OPENROUTER_API_KEY",
        "ghcr.io/jrapakko/scree:0.4.1"
      ],
      "env": {
        "OPENROUTER_API_KEY": "<OPENROUTER_API_KEY>"
      }
    }
  }
}

Configuration

  • OPENROUTER_API_KEYOpenRouter API key, used to embed the notes when indexing and each query when searching.Required

Tools this server provides (1)

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

search_docs
Searches indexed markdown and returns hits with paths and line ranges.

What it does

Heading-level chunks
Splits markdown by heading so hits are short sections.
Hybrid retrieval
Combines vector search with an exact-word keyword index.
Line-range hits
Each hit has a path, heading trail and lines to read.
Local index
Stores everything in a SQLite file per folder; the index can be rebuilt.
Doctor command
Lists files that would be indexed or ignored and estimates cost.
Repo opt-in
Indexes only repos whose .mcp.json names the server.

Good for

  • Find the design doc section that answers a question worded differently
  • Search ADRs and notes without loading whole files
  • Index a notes folder outside any repo
  • Look up an identifier mentioned in docs

Pricing

Open source

Free; embedding calls billed by OpenRouter

Prices change. Check scree.jrpko.dev before you buy.

scree FAQ

Does it index code?

No, only markdown.

What does it need?

An OpenRouter API key for embeddings, saved with scree key.

What does indexing cost?

About $0.0025 per MB of markdown, once; only changed files are re-embedded.

What is the licence?

MIT.

Is scree free?

Yes. scree is open source, so you can use and self-host it for free.

How do I add scree to Claude Code?

Run this command in your terminal: claude mcp add scree --env OPENROUTER_API_KEY='<OPENROUTER_API_KEY>' -- docker run -i --rm -e OPENROUTER_API_KEY ghcr.io/jrapakko/scree:0.4.1

What are the best alternatives to scree?

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

Alternatives to scree

  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