MCP server directory

Brave Search MCP server

One of the ways to connect Brave Search

Web, news, image and video search over Brave's independent index, served as a hosted MCP endpoint.

Other ways to connect
  • MCP (community)

The Brave Search MCP server provides access to the Brave Search API, allowing agents to perform searches and retrieve relevant information for various use cases. It offers features like LLM Context for machine-readable results, Goggles for re-ranking, and extra snippets for riche

Version:
hosted
Publisher:
Brave
Company:
Brave
Docs:
brave.com/search/api/dev-docs
Pricing:
Up to $20 in free calls every month. Prepaid credits, no subscription.
Summary written by google/gemini-2.5-flash-lite from the server’s own site and README; facts above come from the registries.

Why you would use it

A search tool the agent can call without a browser, backed by an index that is not a rebrand of someone else's. Good default when a task starts with "find out".

Brave's search server exposes web search and local search as MCP tools. An agent asks a question, gets ranked results with titles and snippets, and decides which page to open next. No browser session, no scraping, and a rate limit you set with the key you bring.

Install

Generated from the server’s transport data — never hand-written. Placeholders in angle brackets are yours to fill.

Run it with its own identity, a budget you set, and every call audited — no local config to keep in sync.

This server needs credentials — you will be asked for them after it is added.

Add to a connect0 project

You will sign in first, then land back on this page.

What it can do

As declared by the publisher

Tools · 8

  • brave_image_search

    Performs an image search using the Brave Search API. Helpful when you need pictures of people, places, things, graphic-design ideas, or art inspiration. Counts of up to 100 are supported.

  • brave_llm_context

    Retrieves pre-extracted, relevance-ranked web content using Brave's LLM Context API, optimized for AI agents, LLM grounding, and RAG pipelines. Unlike a traditional web search that returns links and short descriptions, this tool returns the actual substance of matching pages — text chunks, tables, code blocks, and structured data — so the model can reason over it directly. When relaying results in markdown-supporting environments, cite source URLs from the `sources` map.

  • brave_local_search

    Brave Local Search API returns enriched information (address, phone, hours, rating) for location-search results. Access requires the Brave Search API Pro plan; currently US-only. Two-step flow: first call `brave_web_search` with `result_filter=locations` to obtain `locations.results[].id`, then pass them here. NOTE: This tool takes location IDs from a prior web-search response; if you have a free-text query, call `brave_web_search` first.

  • brave_news_search

    Searches for news articles using Brave's News Search API. Use it when you need current news, breaking updates, or articles about specific topics.

  • brave_place_search

    Searches Brave's Place Search API. A single call may populate any combination of `results` (POIs), `cities`, `addresses`, `streets`, and `location` (the resolved search area), depending on the query shape. Use for POIs near coordinates or a named area, browsing general POIs (omit `query`, supply coords or `location`), disambiguating bare city names, resolving specific addresses, or looking up streets. Anchor via `latitude`+`longitude` or `location`; with neither, `query` is required.

  • brave_summarizer

    Retrieves AI-generated summaries of web search results. Two-step flow: first call `brave_web_search` with `summary=true` to obtain `summarizer.key`, then pass it here. Pro AI tier required.

  • brave_video_search

    Searches for videos using Brave's Video Search API. Returns structured video results with metadata.

  • brave_web_search

    Performs web searches using the Brave Search API and returns comprehensive search results with rich metadata. To chain into local-POI enrichment, pass `result_filter=locations` and feed the resulting `locations.results[].id` values into `brave_local_search`. To chain into the AI summarizer, pass `summary=true` and feed the returned `summarizer.key` into `brave_summarizer`.

Setup gotchas

  • The hosted endpoint takes a Brave Search API key as a request header; a free key covers a few thousand calls a month.
  • Results are ranked web pages, not answers — pair it with a fetch tool when the agent needs the page body.