Web search for AI agents. Ranked results with page passages already extracted, plus URL to markdown.
com.neurobird/search MCP Server
The com.neurobird/search MCP server provides web search for AI agents. It returns ranked results with pre-extracted page passages, and includes a URL to markdown. It is exposed as an MCP server with 2 tools, described as a search API and listed under the Model Context Protocol.
🛠️ Key Features
Web search for AI agents
Ranked search results
Page passages already extracted
URL to markdown output
2 tools available
🚀 Use Cases
Retrieval for agent workflows using websearch results
Agentic browsing where passages are needed for downstream RAG
Searching for information related to “claude” and LLM-based systems (topics listed)
⚡ Developer Benefits
Structured search output suitable for RAG-style consumption
Markdown-linked results to integrate with text pipelines
MCP compatibility (topics include mcp and model-context-protocol)
⚠️ Limitations
Capabilities described here are limited to web search; no other tools or functions are specified beyond 2 tools and the search output format.
Search the live web and get ranked results with the relevant passages already extracted from each page. Use search_depth 'advanced' for research questions, 'basic' for a quick lookup. Set include_answer to get a grounded summary whose quotes are checked against their sources.
Parameters8
query
string
required
The search query.
search_depth
string
optional
How much work to do. advanced expands the query and reads more pages.
max_results
integer
optional
include_answer
boolean
optional
include_domains
array
optional
Only return results from these domains.
exclude_domains
array
optional
days
integer
optional
Restrict to the last N days.
topic
string
optional
Raw schema
{
"type": "object",
"properties": {
"query": {
"type": "string",
"description": "The search query."
},
"search_depth": {
"type": "string",
"enum": [
"basic",
"standard",
"advanced"
],
"default": "standard",
"description": "How much work to do. advanced expands the query and reads more pages."
},
"max_results": {
"type": "integer",
"default": 6,
"minimum": 1,
"maximum": 20
},
"include_answer": {
"type": "boolean",
"default": true
},
"include_domains": {
"type": "array",
"items": {
"type": "string"
},
"description": "Only return results from these domains."
},
"exclude_domains": {
"type": "array",
"items": {
"type": "string"
}
},
"days": {
"type": "integer",
"description": "Restrict to the last N days."
},
"topic": {
"type": "string",
"enum": [
"general",
"news",
"code",
"science",
"finance"
],
"default": "general"
}
},
"required": [
"query"
]
}
extract_url
Fetch one or more URLs and return their main content as clean markdown, with navigation, ads and boilerplate removed. Pass a query to get only the passages of each page relevant to it.
Parameters3
urls
array
required
URLs to read.
query
string
optional
Optional. Trims each page to its relevant passages.
A hosted Model Context Protocol server that gives an
AI agent web search with the page content already extracted, plus clean markdown
extraction of any URL.
There is nothing to install and nothing to run locally. The server is hosted at
https://search.neurobird.com/mcp over the streamable HTTP transport.
Tested with Claude Code. Cursor, Windsurf, Zed and the Claude desktop app all accept a
streamable HTTP MCP server in this shape.
What makes the results different
Page passages, not snippets. Each result carries the part of the page that answers
the query, extracted from the live page, at no extra credit.
Quotes checked against the source. When an answer is requested, every supporting
quote is matched back against the extracted source text. A quote that fails is returned
marked unverified rather than silently dropped, so an agent can audit a claim instead of
trusting it.
Rank fusion across engines. Results are fused from several engines with reciprocal
rank fusion, so agreement between independent rankings drives the ordering.
Per stage timings in every response, so latency can be attributed from outside.
Pricing
Credits: basic or standard search 1, advanced search 2, grounded answer +1,
extract_url 1 per page.
Per 1,000 searches
Free tier
1,000 credits a month, no card
Basic search
$2
Deep search
$4
Deep search with grounded answer
$6
At 1M credits a month
$1 per 1,000
Honest limits
Median latency on the current single server deployment is about 15 seconds basic,
22 seconds standard and 28 seconds advanced. Upstream engines and the answer model
account for nearly all of it. The HTTP streaming endpoint returns ranked links at
about 5 seconds.
There is no index of our own. Coverage and freshness inherit from upstream engines.
Retrieval capacity, not code, is the scaling limit today.