MCP Deep Research is a tool that allows you to search the web for information. It is built with the [Model Context
MCP Deep Research becomes available to MCP clients through the mcp deep research mcp server. MCP Deep Research is a tool that allows you to search the web for information. It is built with the Model Context Protocol and the Tavily API.
MCP Deep Research is a tool that allows you to search the web for information. It is built with the Model Context Protocol and the Tavily API.
Once connected, the assistant can call these 3 tools directly:
TAVILY_API_KEY — The API key for the Tavily APIMAX_SEARCH_KEYWORDS — The maximum number of search keywords to useMAX_PLANNING_ROUNDS — The maximum number of planning rounds to useThe server is distributed via npm as @smithery/cli, so most clients can run it without a manual build step. Add it to your MCP client's configuration and restart the client to pick it up — the copy-paste configs for Claude Desktop, Claude Code and Cursor are on this page.
Before the server will start you need to supply 3 environment variables: TAVILY_API_KEY, MAX_SEARCH_KEYWORDS, MAX_PLANNING_ROUNDS. Keep credentials in your client's env block or a secrets manager rather than committing them.
Developer-tool servers are usually the first ones people connect, because they turn "help me with this code" into an assistant that can actually read the repo and act on it. MCP Deep Research sits in that group, and the shape of its toolset — TAVILY_API_KEY, MAX_SEARCH_KEYWORDS, MAX_PLANNING_ROUNDS — tells you what it is really for. Worth comparing against the other developer tools servers in this directory before you commit to one, since several overlap in scope but differ sharply in setup cost and permissions.
| Tool | What it does |
|---|---|
| TAVILY_API_KEY | The API key for the [Tavily](https://tavily.com/) API. |
| MAX_SEARCH_KEYWORDS | The maximum number of search keywords to use. |
| MAX_PLANNING_ROUNDS | The maximum number of planning rounds to use. |
{
"mcpServers": {
"deep-research": {
"command": "npx",
"args": ["-y", "mcp-deep-research@latest"],
"env": {
"TAVILY_API_KEY": "your_tavily_api_key", // Required
"MAX_SEARCH_KEYWORDS": "5", // Optional, default 5
"MAX_PLANNING_ROUNDS": "5" // Optional, default 5
}
}
}
}Configuration as documented by the project. Restart the client after saving.
| Variable | Description | Required |
|---|---|---|
| TAVILY_API_KEY | Credential the server authenticates with. | Yes |
| MAX_SEARCH_KEYWORDS | Credential the server authenticates with. | Yes |
| MAX_PLANNING_ROUNDS | Configuration value read at startup. | Optional |
Kill hallucinated APIs — version-accurate, up-to-date library documentation injected straight into context.
Microsoft's official browser automation server — drive a real browser through the accessibility tree, no screenshots needed.
GitHub's official server — repos, issues, pull requests, Actions and code security, straight from your assistant.
Issue tracking at the speed of conversation — Linear's official hosted server with OAuth and zero install.
Local repository surgery — status, diffs, commits, branches and history for any repo on disk.
Timezone sanity for AI — current time anywhere and correct conversions, without the model doing date math.