Developer

The Developer MCP Server helps your team preserve and navigate the entire development context — including projects, components, features, issues, tasks, dependencies, decisions, and milestones — across coding sessions. It keeps you aligned, ready to resume where you left off, and provides a centralized view of project progress and relationships.

How to use

You interact with the Developer MCP Server through an MCP client to manage and query your development context. Start new sessions to record what you accomplish, load detailed context for projects or components, and end sessions to persist updates into a knowledge graph. Use the server to discover related entities, analyze decisions, and track milestone progress. For day-to-day work, you can load the current status of a project, see what tasks are high priority, and quickly navigate dependencies so you can make informed changes without losing context.

How to install

Prerequisites: have Node.js and npm installed on your machine. Ensure you can run npm commands from your shell.

Step 1: Clone the repository.

git clone https://github.com/tejpalvirk/contextmanager.git

Step 2: Install dependencies.

cd contextmanager

npm install

Step 3: Build the server.

npm run build

Step 4: Run the Developer MCP Server.

cd developer

node developer_index.js

Configuration and running with MCP clients

You can configure where the server stores its memory graph data and session data through environment variables when you run it. The following commands show how to set custom paths for memory and session data, while launching the MCP server with npx.

The environment variables shown are MEMORY_FILE_PATH and SESSIONS_FILE_PATH.

Claude Desktop configuration

{
  "mcpServers": {
    "developer": {
      "command": "npx",
      "args": [
        "-y",
        "github:tejpalvirk/contextmanager-developer"
      ]
    }
  }
}

Docker and alternative run methods

If you prefer to run via Docker or another runtime, you can adapt the commands shown in the same configuration style. Ensure the command, arguments, and any necessary environment variables are supplied in your chosen MCP client configuration.

Building and running from source with steps recap

To summarize the build-and-run flow from source: clone the repository, install dependencies, build the server, and finally run the server from the developer module.

Notes on environment variables and usage examples

Examples show how to set data file paths and start the server in the current directory, a specific absolute path, or in your home directory. These examples illustrate how to customize where the server stores its persistent memory and session data.

License

This MCP server is licensed under the MIT License. This means you are free to use, modify, and distribute the software, subject to the terms and conditions of the MIT License. For more details, please see the LICENSE file in the project.

Usage tips and capabilities

Use the MCP server to start new sessions, load detailed context for any entity, and end sessions with structured summaries and task updates. Create new entities and relationships as work progresses, and query decisions, milestones, and dependencies to maintain project alignment across your team.

Available tools

startsession

Starts a new development session and provides information about recent sessions, active projects, high-priority tasks, and upcoming milestones.

loadcontext

Loads detailed context for an entity (project, component, feature, task, etc.) and tracks this context loading as part of the current session.

endsession

Performs a structured analysis of the development session through multiple stages and records this information in the persistent knowledge graph.

buildcontext

Creates new entities, relations, or observations in the knowledge graph.

deletecontext

Removes entities, relations, or observations from the knowledge graph.

advancedcontext

Retrieves information from the knowledge graph with different query types (graph, search, nodes, related, decisions, milestone).