slivingdoc

slivingdoc

A distributed durable notebook MCP server for high-scale agents, providing git-like pull and commit tools with automatic conflict resolution.

Category
Visit Server

README

Test coverage: 83.7% 😍👌

slivingdoc banner

<div align="center"> <p>Distributed durable notebook built for high scale agents.</p> </div>

Features

  • Gitlike semantics: slivingdoc uses terminology we (and agents) all know, designed for ease of use
  • Automatic conflict resolution: commit without fear, trust that all conflicts must be resolved before being accepted
  • High speed processing: the solution is quite simple conceptually, allowing for very high scale and parallelism
  • Plug-and-play: setup the bucket, point at it, and start syncing notes!

docs/slivingdoc-v1.md is the full accepted contract behind these guarantees.

Get started

Add it to your agentic harness:

{
  "mcpServers": {
    "slivingdoc": {
      "command": "npx",
      "args": [
        "-y",
        "slivingdoc",
        "serve",
        "--bucket",
        "my-notes",
        "--workspace-root",
        "/srv/notes"
      ],
      "env": {
        "AWS_ACCESS_KEY_ID": "<your-access-key-id>",
        "AWS_SECRET_ACCESS_KEY": "<your-secret-access-key>",
        "AWS_REGION": "us-east-1"
      }
    }
  }
}

The bucket must exist, and credentials come from the normal AWS chain. No S3 account yet? examples/seaweedfs/ runs a local SeaweedFS container with a step-by-step walkthrough. You can also download a native binary directly from the GitHub release (slivingdoc-v<semver>-<os>-<arch>) and run it in place.

Supported platforms: Linux (amd64, 32-bit ARMv7, arm64), macOS (amd64, arm64), and Windows (amd64). The 32-bit Linux ARM artifact supports Raspberry Pi OS armhf.

How it works

The server exposes two MCP tools over stdio:

Tool Inputs Success result
notes_pull path OK
notes_commit path, message OK

notes_pull writes the current notebook into your directory. notes_commit publishes your changes and incorporates concurrent non-conflicting changes. Between calls there is no protocol at all — agents edit the files with the tools they already have, and humans can write in the same directory with any editor. The next commit carries their changes too.

Humans can also drive both operations directly, without an MCP host. A relative path resolves against the working directory:

slivingdoc pull notes
# edit UTF-8 text files under notes/
slivingdoc commit notes -m "meeting summary"

Success prints the unified result report: the OK status, the accepted remote generation, per-file insertion and deletion counts, and a totals trailer. A domain error exits nonzero and prints a candid report: the error category, the retryable verdict, and every conflicted file with its line ranges. Colour appears only on a real terminal and is disabled by any non-empty NO_COLOR.

The git part

Letting all agents write at once would work, but they would get overrun by race conditions. So slivingdoc has built-in git via libgit2 which effectively does:

  1. git pull
  2. (potential conflict resolution locally)
  3. git add .
  4. git commit -m "<agent message>"
  5. git push
  6. (potential conflict resolution locally)

All of these git operations are handled locally within a private mirror of the notes directory leaving a "streamlined" git sequence. This works due to two compromises, firstly that the local notes directory is prone to be changed on notes_pull, precedence goes to the remote state, leaving conflict markers. Secondly, the system only works for text (clean UTF-8).

Configuration

serve, pull, and commit read the same flags and environment variables. --bucket is required. The most common flags:

Flag Environment Default
--bucket SLIVINGDOC_BUCKET — (required)
--workspace-root SLIVINGDOC_WORKSPACE_ROOT startup working dir
--endpoint AWS_ENDPOINT_URL_S3 AWS resolution
--region AWS_REGION us-east-1

slivingdoc serve -h prints the full reference, and docs/running.md covers everything an operator needs: all flags, the exact S3 permissions, logging (LOG_LEVEL on stderr), the notebook rules, conflict recovery, and checkpoint retention. The Terraform module in terraform/ provisions a bucket and a least-privilege IAM user for one notebook.

Development

  • AGENTS.md — the developer and agent guide: package map, operation flows, conventions, and the QA gates.
  • docs/slivingdoc-v1.md — the accepted architecture contract.
  • docs/build.md — the native build, from pinned libgit2 source to dependency inspection.
  • docs/testing.md — the test commands, the test layers, and the no-live-AWS rule.
  • docs/releasing.md — release artifacts, npm trusted publishing, and make release.
make qa #lint plus the full Go and npm test suites

License

MIT — see LICENSE. Third-party notices for the statically linked libgit2 are in NOTICE.

<sub>Not affiliated with Paris Hilton.</sub>

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured