# ContextStream MCP server

> Give AI coding assistants persistent memory, code search and shared project context that carry across sessions and tools.

- Listing: https://mcp.tc/i/contextstream
- Connect: use the server's own URL `https://mcp.contextstream.io/mcp?default_context_mode=fast` (OAuth sign-in at the server); clients connect to it directly. The listing link is a page, not an MCP endpoint.
- Type: remote (Streamable HTTP)
- Auth: OAuth sign-in
- Category: [AI & Machine Learning](https://mcp.tc/c/ai-ml)
- Vendor: ContextStream
- Homepage: <https://contextstream.io/docs/mcp>
- Repository: <https://github.com/contextstream/mcp-server>
- Package: npm `@contextstream/mcp-server`

## About

ContextStream connects your code, docs and conversations to AI coding assistants. Agents can search indexed code, save decisions and lessons, and retrieve them in later sessions with traceable sources and scoped access.

It runs as a hosted Streamable HTTP endpoint with OAuth sign-in, or as a local launcher installed with npx (Node.js 20+) that downloads a native binary. Indexing sends eligible source contents to ContextStream for hosted search, and transcript saving and local Git metadata capture are on by default and can be turned off. A free tier with monthly credits is offered.

## What it can do

- Save decisions and lessons and recall them in later sessions
- Search indexed code and docs with cited sources
- Share project context across editors and agents
- Work with clients such as Claude Code, Cursor, Codex and Gemini CLI
- Control transcript saving and Git metadata capture

## Example prompts

- "Use ContextStream to find the files relevant to my next change."
- "Save this decision and its reason to project memory."
- "Retrieve the saved decisions that should guide this refactor."
- "Search my indexed project for where authentication is handled."

## Install

### Claude Code

1. Run this in a terminal, in your project folder:

```bash
claude mcp add --transport http contextstream 'https://mcp.contextstream.io/mcp?default_context_mode=fast'
```

2. Start Claude Code, type `/mcp`, pick **contextstream** and choose **Authenticate**. A browser window opens for the ContextStream sign-in.

Add `--scope user` to make it available in every project, not just this one.

### Claude Desktop

1. Open **Customize → Connectors**, click **+ Add**, then **Add custom connector**.

2. Name it **ContextStream** and paste this URL:

```url
https://mcp.contextstream.io/mcp?default_context_mode=fast
```

3. Click **Add**, then **Connect**, and sign in when ContextStream asks.

Claude Desktop’s JSON config file only starts local servers. Remote servers go through Connectors, and connectors you add on claude.ai show up here too.

### claude.ai

1. Open the connector form on claude.ai. This button fills in the name and URL for you:

[Add to claude.ai](<https://claude.ai/customize/connectors?modal=add-custom-connector&connectorName=ContextStream&connectorUrl=https%3A%2F%2Fmcp.contextstream.io%2Fmcp%3Fdefault_context_mode%3Dfast>) (opens connector settings)

2. Check that the URL reads `https://mcp.contextstream.io/mcp?default_context_mode=fast` and click **Add**.

3. Click **Connect** and sign in when ContextStream asks.

Free plans allow one custom connector. On Team and Enterprise plans an owner adds it under **Organization settings → Connectors**.

### ChatGPT

1. On chatgpt.com, go to **Plugins**, select **+**, then **Add custom MCP server**.

2. Enter a name for the server.

3. Under **Connection**, paste `https://mcp.contextstream.io/mcp?default_context_mode=fast` as the server URL, then for authentication choose **OAuth**. ChatGPT sends you to ContextStream to sign in.

4. Check **I understand and want to continue**, then select **Create as a plugin**.

Whether you can add a custom MCP server depends on your ChatGPT plan and your workspace permissions (in Business workspaces, admins and owners can).

### Cursor

[Add to Cursor](<https://cursor.com/install-mcp?name=contextstream&config=eyJ1cmwiOiJodHRwczovL21jcC5jb250ZXh0c3RyZWFtLmlvL21jcD9kZWZhdWx0X2NvbnRleHRfbW9kZT1mYXN0In0%3D>) (opens Cursor)

Or add it by hand to `~/.cursor/mcp.json` (all projects) or `.cursor/mcp.json` (this project):

`mcp.json`:

```json
{
  "mcpServers": {
    "contextstream": {
      "url": "https://mcp.contextstream.io/mcp?default_context_mode=fast"
    }
  }
}
```

Cursor shows **Needs authentication** next to the server. Click the button beside it to sign in.

### VS Code

[Install in VS Code](<https://vscode.dev/redirect/mcp/install?name=contextstream&config=%7B%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fmcp.contextstream.io%2Fmcp%3Fdefault_context_mode%3Dfast%22%7D>) (opens VS Code)

Or from a terminal:

```bash
code --add-mcp '{"name":"contextstream","type":"http","url":"https://mcp.contextstream.io/mcp?default_context_mode=fast"}'
```

Or commit it to the repo in `.vscode/mcp.json`:

`.vscode/mcp.json`:

```json
{
  "servers": {
    "contextstream": {
      "type": "http",
      "url": "https://mcp.contextstream.io/mcp?default_context_mode=fast"
    }
  }
}
```

VS Code asks you to sign in the first time the server starts.

### Devin Desktop

1. Add it to `~/.config/devin/mcp_config.json` (macOS and Linux) or `%APPDATA%\devin\mcp_config.json` (Windows):

`mcp_config.json`:

```json
{
  "mcpServers": {
    "contextstream": {
      "serverUrl": "https://mcp.contextstream.io/mcp?default_context_mode=fast"
    }
  }
}
```

2. Refresh the MCP server list in Cascade and sign in when asked.

Devin Desktop is the new name for Windsurf. It reads `serverUrl` (or `url`) for remote servers.

### Codex

```bash
codex mcp add contextstream --url 'https://mcp.contextstream.io/mcp?default_context_mode=fast'
codex mcp login contextstream
```

Or edit `~/.codex/config.toml` directly:

`config.toml`:

```toml
[mcp_servers.contextstream]
url = "https://mcp.contextstream.io/mcp?default_context_mode=fast"
```

### Gemini CLI

```bash
gemini mcp add --transport http contextstream 'https://mcp.contextstream.io/mcp?default_context_mode=fast'
```

Then, inside Gemini CLI, run `/mcp auth contextstream` to sign in.

This adds it to the current project. Add `-s user` to use it everywhere.

### Any client

Most clients accept this shape. Some name the URL field differently: `serverUrl` in Devin Desktop, `httpUrl` in Gemini CLI’s settings file.

```json
{
  "mcpServers": {
    "contextstream": {
      "type": "http",
      "url": "https://mcp.contextstream.io/mcp?default_context_mode=fast"
    }
  }
}
```

Zed puts servers under `context_servers` in its settings. Cline needs `"type": "streamableHttp"`, or it assumes SSE.

Client only starts local servers? Bridge it with `npx -y mcp-remote 'https://mcp.contextstream.io/mcp?default_context_mode=fast'`.

## Details

- Server version: 0.4.81
- Last checked: 2026-11-17 (reachable, asks for credentials)
- Listed: 2026-10-04
- Updated: 2026-10-04

---
Source: https://mcp.tc/i/contextstream (mcp.tc is an independent directory, not affiliated with this server's publisher). Corrections: https://mcp.tc/report
