ContextStream

v0.3.63Knowledge & Memorystable

Provides AI assistants with persistent memory and code intelligence across all tools and conversations. Features semantic search, knowledge graphs, decision tracking, and impact analysis with 60+ tools for universal context preservation.

ai-memoryclaudecontext-managementcursormcp
Share:
36
Stars
0
Downloads
0
Weekly
0/5

What is ContextStream?

ContextStream is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to provides ai assistants with persistent memory and code intelligence across all tools and conversations. features semantic search, knowledge graphs, decision tracking, and impact analysis with 60+ tool...

Provides AI assistants with persistent memory and code intelligence across all tools and conversations. Features semantic search, knowledge graphs, decision tracking, and impact analysis with 60+ tools for universal context preservation.

This server falls under the Knowledge & Memory and Developer Tools categories on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Provides AI assistants with persistent memory and code intel

Use Cases

Maintain persistent memory and code intelligence across multiple tools and conversations.
Search and analyze knowledge graphs to track decisions and their impact.
Preserve universal context with 60+ tools for AI assistants across different editors.
contextstream

Maintainer

LicenseMIT License
Languagetypescript
Versionv0.3.63
UpdatedMay 16, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y @contextstream/mcp-server

Manual Installation

npx -y @contextstream/mcp-server

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use ContextStream

ContextStream MCP Server gives AI assistants persistent memory and deep code intelligence that survives across conversations, sessions, and tools. It connects to the ContextStream cloud platform to provide semantic search over your codebase, a project knowledge graph, decision tracking, impact analysis across dependencies, and integration with GitHub, Slack, and Notion — all through 9 core tools accessible from Claude Code, Cursor, VS Code, Windsurf, and other MCP clients. Engineering teams working across large codebases benefit most, as the server preserves context that would otherwise be lost when starting a new chat or switching between AI tools.

Prerequisites

  • Node.js 18+ installed on your system
  • A ContextStream account and API key — sign up at https://contextstream.io to obtain CONTEXTSTREAM_API_KEY
  • An MCP-compatible client such as Claude Desktop, Claude Code, Cursor, or VS Code with MCP support
  • Internet access to reach the ContextStream API at https://api.contextstream.io
1

Run the ContextStream setup wizard

The quickest way to configure ContextStream is via the setup command, which auto-detects your installed MCP clients and adds the server configuration for you.

npx --prefer-online -y @contextstream/mcp-server@latest setup
2

Obtain your ContextStream API key

If you have not already signed up, visit https://contextstream.io and create an account. From your dashboard, copy your API key — you will set it as CONTEXTSTREAM_API_KEY in the config.

3

Manually configure your MCP client (if not using setup wizard)

If the setup wizard did not detect your client, manually add the ContextStream server entry to your claude_desktop_config.json or equivalent config file.

4

Initialize your workspace context

After connecting, use the init tool to index your current workspace. Ask Claude to initialize the ContextStream context for your project — this indexes the codebase for semantic search.

5

Verify tools are available

Ask Claude to list available ContextStream tools or run a semantic search to confirm the server is connected and the workspace is indexed.

ContextStream Examples

Client configuration

Add this entry to your claude_desktop_config.json. Set CONTEXTSTREAM_API_KEY to your key from the ContextStream dashboard. Optionally set CONTEXTSTREAM_TOOLSET to 'complete' for all available tools.

{
  "mcpServers": {
    "contextstream": {
      "command": "npx",
      "args": ["-y", "@contextstream/mcp-server"],
      "env": {
        "CONTEXTSTREAM_API_KEY": "your_api_key_here",
        "CONTEXTSTREAM_API_URL": "https://api.contextstream.io",
        "CONTEXTSTREAM_TOOLSET": "complete"
      }
    }
  }
}

Prompts to try

Use Claude with ContextStream to search code semantically, track decisions, and analyze dependencies.

- "Initialize ContextStream for this workspace and index the codebase."
- "Search the codebase for all places where authentication tokens are validated."
- "Save a session note: we decided to use PostgreSQL instead of MongoDB for reasons X, Y, Z."
- "Show me the dependency graph for the UserService module and what it affects."
- "What are the key decisions recorded for this project?"
- "Search for code that handles payment processing and show related GitHub issues."

Troubleshooting ContextStream

Authentication error: invalid or missing API key

Verify that CONTEXTSTREAM_API_KEY is set correctly in your MCP config env block. The key must come from your ContextStream dashboard at https://contextstream.io. Ensure there are no extra spaces or newlines in the key value.

Search returns no results even after initializing the workspace

Indexing large codebases can take several minutes. Wait a few minutes after running init before querying. Also confirm CONTEXTSTREAM_API_URL is set to https://api.contextstream.io — a wrong URL causes silent empty results.

Tools are not appearing in Claude Desktop after adding the config

Fully quit and relaunch Claude Desktop (not just close the window). Check that the JSON in claude_desktop_config.json is valid — any syntax error silently prevents all MCP servers from loading. Use a JSON validator if unsure.

Frequently Asked Questions about ContextStream

What is ContextStream?

ContextStream is a Model Context Protocol (MCP) server that provides ai assistants with persistent memory and code intelligence across all tools and conversations. features semantic search, knowledge graphs, decision tracking, and impact analysis with 60+ tools for universal context preservation. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install ContextStream?

Install via npm with the command: npx -y @contextstream/mcp-server. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with ContextStream?

ContextStream works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is ContextStream free to use?

Yes, ContextStream is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

Browse More Knowledge & Memory MCP Servers

Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

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

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use ContextStream?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides