Synapse
A project-agnostic MCP server that exposes Markdown documentation from a project's /docs folder as MCP resources for AI agents. It provides stable, up-to-date context to reduce hallucinations and ensure agents remain aligned with project-specific conventions and goals.
README
Synapse
Project-agnostic MCP server that turns your /docs folder into a shared “brain” for AI agents.
Synapse exposes your project’s documentation as Model Context Protocol (MCP) resources, so any MCP-capable client (e.g. Cursor, Claude Desktop) can use them as stable, up-to-date context. The goal is to reduce hallucinations and keep agents aligned with your project’s vision, conventions, and goals.
Why Synapse?
- Agnostic — Works with any repo. No product-specific logic; you bring the docs, Synapse serves them.
- Convention over config — Put Markdown in
/docsat the project root; the server discovers and exposes it. - OSS — MIT license. You can use, change, and contribute.
Status
Early stage. The repo has the basic TypeScript + MCP SDK setup and a stub entrypoint. Next steps:
- Implement MCP resources: list and read all
.mdfiles under/docs(URIs likesynapse://docs/...). - Optional config (e.g. custom docs path, handoffs) later.
Requirements
- Node.js ≥ 18
- pnpm (recommended) or npm
Installation
# Clone the repo
git clone https://github.com/gabriel-romero-poza/synapse-mcp.git
cd synapse-mcp
# Install dependencies
pnpm install
# Build
pnpm build
Or use as a dependency in another project:
pnpm add synapse-mcp # when published to npm
# or from git
pnpm add github:gabriel-romero-poza/synapse-mcp
Usage
Run the server (stdio transport). The project root is the directory from which the MCP client runs the process (e.g. your repo root when Cursor starts Synapse).
pnpm start
# or after npm/pnpm link / global install:
synapse
Environment (optional):
| Variable | Description |
|---|---|
SYNAPSE_PROJECT_ROOT |
Override project root (default: process.cwd()) |
Cursor
- Settings → Tools & MCP → Add new MCP server
- Add a server with:
- Name:
synapse(or any name) - Type:
command - Command:
node - Arguments:
path/to/synapse/dist/index.js
(orpnpm exec synapseif run from the synapse repo / a workspace that has it)
- Name:
- Ensure the server is started with cwd = your project root (where
/docslives).
(Exact Cursor UI may vary; refer to Cursor MCP docs.)
Project layout
synapse-mcp/
├── .cursor/rules/ # Cursor rules (commits, TypeScript, agentic workflow)
├── .github/ # PR template, issue templates
├── docs/ # Project vision and design (source of truth)
├── src/
│ └── index.ts # Entrypoint (MCP server stub)
├── AGENTS.md # Context for AI agents
├── CONTRIBUTING.md # Commit conventions, PR process, guardrails
├── package.json
├── tsconfig.json
├── pnpm-lock.yaml
├── README.md
└── LICENSE
Contributing
Contributions are welcome. See CONTRIBUTING.md for commit conventions (Conventional Commits), PR process, and guardrails. Open an issue for bigger changes; use pull requests for patches and small features. Keep the scope agnostic: no product-specific logic inside Synapse; conventions and optional config only.
License
MIT. See LICENSE.
Recommended Servers
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.
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.
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.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
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.
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.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
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.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.