MCP Research Friend
Research tools, including a Sqlite-backed document stash
Ads
Overview
A friendly helper for AI assistants that need to look things up on the web and manage a local research stash. Research Friend is an MCP server that gives your AI tools the ability to fetch web pages and search the internet. It uses a real web browser behind the scenes, so it works even with modern websites that rely heavily on JavaScript. It also includes a local “stash” for storing documents, extracting text, and searching across your library. To make use of all its features, you'll want an MCP client that supports prompts (common) and sampling (less common). We're building Research Friend alongside Chabeau, which supports both.
Installation
npm install
Configuration
Configuration example is not available yet.
Capabilities
- Fetch and extract web pages
Ask your assistant to retrieve a page’s main content as clean markdown using `friendly_web_fetch`.
- Search the web
Have your assistant run a web search via DuckDuckGo or Google with `friendly_search` and return titles, URLs, and snippets.
- Extract PDFs or paginate large pages
Use `friendly_web_extract` to pull text from a PDF, paginate with offsets, or search for phrases within a document.
- Answer questions about a URL
Ask your assistant to fetch a page or PDF and have an LLM answer a specific question about it via `friendly_web_ask`.
- Build and search a local document stash
Drop files into the inbox, process them with `stash_process_inbox`, then search across filenames and content using `stash_search`.
- Query a stashed document with an LLM
Use `stash_ask` to ask questions about a stored document without loading it into the main conversation.
Tags
- search
- knowledge-retrieval
- web-research
Related MCP servers
Mastra Docs Server
OfficialProvides AI assistants with direct access to Mastra.ai's complete knowledge base.
Fetch MCP Server
OfficialFetch web content and convert HTML pages into LLM-friendly markdown output.
Perplexity Ask MCP Server
OfficialA connector for the Perplexity API to enable web search within the MCP ecosystem.
Perplexity
OfficialAn MCP server that connects to Perplexity's Sonar API, enabling real-time web-wide research in conversational AI.