NervusDB

Provides a code knowledge graph workflow with project indexing, analysis, and code operations for NervusDB.

typescript

6

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "ysankpia-nervusdb-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@nervusdb/mcp"
      ],
      "env": {
        "GITHUB_TOKEN": "YOUR_GITHUB_TOKEN"
      }
    }
  }
}

You run NervusDB’s MCP server to build a cross-language code knowledge graph, analyze code impact, automate workflows, read and write files, and query the knowledge index. This MCP server ties your codebase to a knowledge graph so you can explore relationships, manage tasks, and orchestrate code operations across projects.

How to use

Start by wiring an MCP client to your NervusDB MCP server. Use the provided CLI entry point to launch, index your project, and then call the available tools to analyze, read, write, and query. You can create tasks, push branches, run tests, and submit changes for review, all while keeping the knowledge graph in sync with your repository.

How to install

Prerequisites: install Node.js 20.0.0 or higher and pnpm 8.0.0 or higher.

Install dependencies for the MCP server.

pnpm install

How to use one-click Claude Desktop integration

Configure Claude Desktop to run the MCP server directly from your environment. Add the following to the Claude configuration on macOS to start the MCP server via npx:

{
  "mcpServers": {
    "nervusdb-mcp": {
      "command": "npx",
      "args": ["-y", "@nervusdb/mcp"]
    }
  }
}

Starting the server for development or production

Run the server in development mode or build for production, then start the MCP server accordingly.

# For development
pnpm start:stdio

# Build for production
pnpm build

Indexing a project

Index your project into the knowledge graph to enable analysis, impact assessment, and search across the codebase.

pnpm synapse:index -p /path/to/your/project

Available tools

workflow.startTask

Create a task branch and update the task ledger to track progress across your workflow.

workflow.submitForReview

Push the task branch and submit a pull request for review (requires GitHub authentication).

project.getStructure

Get the project file structure with statistics to understand code layout.

project.analyzeImpact

Analyze code impact based on the knowledge graph to see how changes propagate.

project.findRelatedFiles

Find files related to a target file within the project.

project.readFile

Read the content of an arbitrary file in the project.

code.readFile

Read content of a specific project file.

code.writeFile

Write content to a project file with an explicit confirmation step.

code.runTests

Run tests using Vitest and return the results.

db.getStats

Get index metadata and statistics for the knowledge graph.

db.query

Execute typed or raw queries against the knowledge graph.

db.rebuildIndex

Rebuild the project index with telemetry data.

db.getHealth

Check the health of the index with fingerprint validation.