Web research over MCP with scored Google results, clean webpage extraction, and multi-source topic research.
Copy the AI prompt to install this server into Claude Code, Cursor, or another agent — or use 1-click editor setup below.
💡 Paste the JSON block into your client's configuration file under mcpServers, then restart the application.
Inspect callable tools, capabilities, and parameters exposed to AI agents by Nimrod Research.
Nimrod Research MCP server provides web-research operations through the Model Context Protocol. It connects an MCP client to Google search and web pages, returning structured search evidence or cleaned page content for use by an agent. The service is aimed at research tasks that need current sources, source ranking, and text that is more compact than raw HTML.
Four tools are documented:
google_search returns organic Google results with quality scores, authority ratings, and source-type labels.extract_webpage_content converts one URL into Markdown after removing page boilerplate.extract_multiple_webpages processes as many as five URLs concurrently.research_topic combines search, ranking, deduplication, page extraction, and synthesis in one request.Search costs one credit. Topic research costs one credit for the search plus one credit for each focus area. Webpage extraction is free. New accounts include a 50-credit trial without requiring a payment card.
A hosted MCP endpoint is available at https://nimrod.orchis.ai/mcp. Claude.ai and Claude Desktop can connect through the custom connector settings and an email sign-in. Claude Code can use the same endpoint over HTTP with a Nimrod bearer key.
Nimrod Research MCP server can also run through desktop downloads. The README describes a Claude Desktop package for Windows and executable desktop builds for Windows, macOS, and Linux that can be configured for Claude Code. In that local setup, extraction runs on the user’s machine, so pages read through the desktop tool remain private according to the project’s documentation.
The project is also published in the official MCP Registry under ai.orchis/nimrod, allowing registry-aware clients such as VS Code’s MCP gallery to discover and install it.
For Claude.ai or Claude Desktop, open Settings, choose Connectors, select Add custom connector, enter the hosted MCP URL, and authenticate with an email address.
For Claude Code, add the hosted server with an HTTP transport and an authorization header containing a Nimrod key:
Desktop installation differs by platform. Windows provides an .mcpb package for Claude Desktop and a desktop executable for Claude Code. The macOS binary is universal but requires executable permission and removal of its quarantine attribute before setup. Linux also requires making the downloaded binary executable before running setup. The README does not specify environment variables.
The search tool supports date, site, and language control, in addition to quality and authority information. Single-page extraction is useful when an agent already has a URL; concurrent extraction handles a small group of known sources. Topic research is the higher-level option when the agent should discover, filter, read, and synthesize sources in one call.
Search and topic research consume credits, while extraction does not. Topic-research cost varies with the number of focus areas. The hosted Claude Code example requires a bearer key, and the desktop macOS binary is not Apple-signed. The original open-source BYOK implementation remains on a historical branch marked unmaintained.
Factual signals from GitHub, npm, and our automated checks — not a rating.
No reviews yet — be the first to share how this listing worked for you.
Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.
[](https://allmcps.com/mcp/nimrod-research)<a href="https://allmcps.com/mcp/nimrod-research"><img src="https://allmcps.com/api/badge/nimrod-research?style=directory" alt="Nimrod Research on AllMCPs" /></a>