Context MCP Server

LLM Context Manager MCP Server & CLI wrapper using npx

Local serverstdioPython

What is the Context MCP server?

LLM Context Manager MCP Server & CLI wrapper using npx. The context mcp server wraps that behind the Model Context Protocol, so an assistant can use it through 9 defined tools rather than through you.

What it actually does

Modern AI coding assistants (Claude, Cursor, Windsurf) are powerful but suffer from blindness:

Its toolset

Everything the assistant can do here goes through one of these:

  • search_code — Hybrid semantic + graph search
  • get_context — Cursor-based retrieval
  • find_nodes — Find nodes by name or path
  • read_graph — Inspect a specific node by ID
  • index_project — Refresh the project index
  • find_usages — Find all callers of a node
  • trace_call_chain — BFS call chain between two nodes
  • impact_of_change — Blast radius of a file change
  • diff_context — Recently changed code via git

Adding it to your client

Installation goes through your MCP client rather than a global install: point it at @senoldogann/context-manager on npm and it is fetched when the client starts. The copy-paste blocks for Claude Desktop, Claude Code and Cursor are further down this page.

Configuration

You will need 4 environment variables: EMBEDDING_HOST, EMBEDDING_API_KEY, CCM_DB_PATH, OPENAI_API_KEY. The server will not start without them, which is usually why the tools fail to appear on a first run. Keep credentials in your client's env block or a secrets manager rather than in a file you might commit.

When to reach for it

Plenty of developer tooling servers cover similar ground. The differences that matter in practice are scope of access and how much setup stands between you and a working tool call. Context's toolset — search_code, get_context, find_nodes and 6 more — is a fair guide to whether it matches your workflow. It is maintained by senoldogann; worth a glance at recent repository activity before you build anything load-bearing on it.

SyncDev reviews every entry in this directory against the project's own documentation before publishing, and revisits them as servers change.

Caveats

  • It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
  • With 9 tools registered it takes up a noticeable share of the context window; turn it off in projects that never touch Context.
  • Missing credentials fail quietly in some clients — if no tools show up, check the environment block first.
  • MCP clients confirm each tool call by default. Leave that on until you have watched what the context mcp server does with a few real requests.

Available tools

ToolWhat it does
search_codeHybrid semantic + graph search
get_contextCursor-based retrieval
find_nodesFind nodes by name or path
read_graphInspect a specific node by ID
index_projectRefresh the project index
find_usagesFind all callers of a node
trace_call_chainBFS call chain between two nodes
impact_of_changeBlast radius of a file change
diff_contextRecently changed code via git

How to install the Context MCP server

{
  "mcpServers": {
    "context-manager": {
      "command": "npx",
      "args": ["-y", "@senoldogann/context-manager"],
      "env": {
        "EMBEDDING_HOST": "your-value",
        "EMBEDDING_API_KEY": "your-value",
        "CCM_DB_PATH": "your-value",
        "OPENAI_API_KEY": "your-value"
      }
    }
  }
}

Add to claude_desktop_config.json, then restart Claude Desktop.

Configuration

VariableDescriptionRequired
EMBEDDING_HOSTEndpoint or connection string the server talks to.Optional
EMBEDDING_API_KEYCredential the server authenticates with.Yes
CCM_DB_PATHFilesystem location the server is allowed to use.Optional
OPENAI_API_KEYCredential the server authenticates with.Yes

Example prompts to try

  • Use Context to search code.
  • Use Context to get context.
  • Use Context to find nodes.

Frequently asked questions

It connects Context to MCP-compatible AI assistants such as Claude and Cursor, exposing 9 tools (search_code, get_context, find_nodes, and more) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Context directly.