How do I use google-research-mcp?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@google-research-mcp research the latest advances in quantum computing" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

google-research-mcp

by thejusdutt

Overview Schema Source Related Servers Discussions Admin

TypeScript

Remote

Google Research MCP Server v2.0.0 - Multi-Agent Architecture

An MCP server that implements Anthropic's Multi-Agent Research Architecture with true subagent spawning, adaptive stopping, and citation processing.

Architecture Overview

This implementation is fully compliant with Anthropic's multi-agent research system:

┌─────────────────────────────────────────────────────────────────┐
│                    Multi-Agent Research System                   │
├─────────────────────────────────────────────────────────────────┤
│                                                                  │
│  ┌──────────────────────────────────────────────────────────┐   │
│  │              LEAD RESEARCHER (Orchestrator)               │   │
│  │                                                           │   │
│  │  • think(plan approach) - Decompose into aspects          │   │
│  │  • create subagents - Spawn parallel workers              │   │
│  │  • think(synthesize) - Combine findings                   │   │
│  │  • evaluate coverage - "More research needed?"            │   │
│  │  • complete_task - Return final report                    │   │
│  └──────────────────────────────────────────────────────────┘   │
│                              │                                   │
│              ┌───────────────┼───────────────┐                  │
│              ▼               ▼               ▼                  │
│  ┌────────────────┐ ┌────────────────┐ ┌────────────────┐      │
│  │  SUBAGENT 1    │ │  SUBAGENT 2    │ │  SUBAGENT N    │      │
│  │  (Aspect A)    │ │  (Aspect B)    │ │  (Aspect N)    │      │
│  │                │ │                │ │                │      │
│  │ • web_search   │ │ • web_search   │ │ • web_search   │      │
│  │ • think(eval)  │ │ • think(eval)  │ │ • think(eval)  │      │
│  │ • complete     │ │ • complete     │ │ • complete     │      │
│  └────────────────┘ └────────────────┘ └────────────────┘      │
│              │               │               │                  │
│              └───────────────┼───────────────┘                  │
│                              ▼                                   │
│  ┌──────────────────────────────────────────────────────────┐   │
│  │                    CITATION AGENT                         │   │
│  │  • Process documents                                      │   │
│  │  • Identify citation locations                            │   │
│  │  • Insert inline citations [1], [2], etc.                │   │
│  │  • Generate references section                            │   │
│  └──────────────────────────────────────────────────────────┘   │
│                              │                                   │
│  ┌──────────────────────────────────────────────────────────┐   │
│  │                    MEMORY MODULE                          │   │
│  │  • save plan                                              │   │
│  │  • retrieve context                                       │   │
│  │  • persist findings                                       │   │
│  │  • track gaps                                             │   │
│  └──────────────────────────────────────────────────────────┘   │
│                                                                  │
└─────────────────────────────────────────────────────────────────┘

Related MCP server: PapersFlow

Process Flow

Based on Anthropic's sequence diagram:

User                LeadResearcher        Subagent1         Subagent2         Memory          CitationAgent
 │                       │                    │                 │                │                  │
 │──send user query────▶│                    │                 │                │                  │
 │                       │                    │                 │                │                  │
 │                       │◀─────────────────────────────────────────────────────│                  │
 │                       │  think(plan approach)                                │                  │
 │                       │                    │                 │                │                  │
 │                       │──save plan────────────────────────────────────────▶│                  │
 │                       │                    │                 │                │                  │
 │                       │──retrieve context──────────────────────────────────▶│                  │
 │                       │                    │                 │                │                  │
 │                       │                    │                 │                │                  │
 │                       │══════════════════════════════════════════════════════│                  │
 │                       │                 ITERATIVE RESEARCH LOOP              │                  │
 │                       │══════════════════════════════════════════════════════│                  │
 │                       │                    │                 │                │                  │
 │                       │──create subagent──▶│                 │                │                  │
 │                       │──create subagent────────────────────▶│                │                  │
 │                       │                    │                 │                │                  │
 │                       │                    │──web_search────▶│                │                  │
 │                       │                    │◀───results──────│                │                  │
 │                       │                    │                 │                │                  │
 │                       │                    │  think(evaluate)│                │                  │
 │                       │                    │                 │                │                  │
 │                       │◀──complete_task────│                 │                │                  │
 │                       │                    │                 │                │                  │
 │                       │                    │                 │──web_search───▶│                  │
 │                       │                    │                 │◀──results──────│                  │
 │                       │                    │                 │                │                  │
 │                       │                    │                 │ think(evaluate)│                  │
 │                       │                    │                 │                │                  │
 │                       │◀─────────────────────complete_task───│                │                  │
 │                       │                    │                 │                │                  │
 │                       │  think(synthesize results)           │                │                  │
 │                       │                    │                 │                │                  │
 │                       │         ┌─────────────────────┐      │                │                  │
 │                       │         │ More research needed?│      │                │                  │
 │                       │         └─────────────────────┘      │                │                  │
 │                       │              │           │           │                │                  │
 │                       │         [Continue]   [Exit Loop]     │                │                  │
 │                       │              │           │           │                │                  │
 │                       │══════════════════════════════════════════════════════│                  │
 │                       │                    │                 │                │                  │
 │                       │──complete_task (research result)────────────────────▶│                  │
 │                       │                    │                 │                │                  │
 │                       │                    │                 │                │──────────────────▶│
 │                       │                    │                 │                │  Process docs +   │
 │                       │                    │                 │                │  insert citations │
 │                       │◀───────────────────────────────────────────────────────────────────────│
 │                       │                    │                 │                │  Report with      │
 │                       │                    │                 │                │  citations        │
 │                       │──persist results──────────────────────────────────▶│                  │
 │                       │                    │                 │                │                  │
 │◀──return research─────│                    │                 │                │                  │
 │   results with        │                    │                 │                │                  │
 │   citations           │                    │                 │                │                  │

Key Features

1. True Subagent Spawning

Each aspect gets its own subagent that runs independently:

Generates aspect-specific queries
Executes web searches
Fetches full page content
Evaluates findings
Reports back to Lead Researcher

2. Think/Evaluate Phases

Explicit reasoning phases between iterations:

think(plan approach) - Decompose topic into aspects
think(evaluate) - Each subagent evaluates its findings
think(synthesize) - Lead Researcher combines all findings

3. Adaptive Stopping

Dynamic "More research needed?" decision:

Coverage score calculation (0-100%)
Configurable thresholds per depth level
Gap identification and filling
Exits early when coverage is sufficient

4. Aspect-Based Decomposition

Topics are broken into researchable aspects:

Basic: 2 aspects (overview, mechanism)
Moderate: 5 aspects (+use cases, benefits, challenges)
Comprehensive: 11 aspects (+history, comparisons, implementation, future, research, case studies)

5. Memory Module

Persistent context across iterations:

Research plan storage
Findings per aspect
Gap tracking
Iteration history

6. Citation Agent

Dedicated citation processing:

Assigns citation IDs by quality
Inserts inline citations [1], [2]
Generates references section
Groups by quality tier

Tools

Tool	Description
`google_research`	Full multi-agent research with all components
`deep_search`	Search + fetch full content (single iteration)
`deep_search_news`	News-specific deep search
`fetch_page`	Fetch single page content
`google_search`	Simple search (snippets only)
`web_search`	Search with quality scoring
`research_session`	Manual session management
`run_subagent`	Manually spawn a subagent
`evaluate_coverage`	Check coverage and gaps
`add_source`	Add source to session
`get_citations`	Format citations

Installation

{
  "mcpServers": {
    "google-research": {
      "command": "npx",
      "args": ["google-research-mcp"],
      "env": {
        "GOOGLE_API_KEY": "your-api-key",
        "GOOGLE_CX": "your-search-engine-id"
      }
    }
  }
}

Prerequisites

1. Google API Key

Go to Google Cloud Console
Enable "Custom Search API"
Create an API Key

2. Search Engine ID (CX)

Go to Programmable Search Engine
Create engine with "Search the entire web"
Copy the Search Engine ID

Usage Examples

Full Multi-Agent Research

"Research quantum computing with comprehensive depth"

This triggers the full architecture:

Lead Researcher plans 11 aspects
Spawns 3-4 subagents per iteration
Each subagent researches in parallel
Evaluates coverage after each iteration
Continues until 90% coverage or max iterations
Citation Agent processes final report

Manual Subagent Control

// Create session
research_session({ action: "create", topic: "AI safety", depth: "moderate" })

// Spawn specific subagents
run_subagent({ sessionId: "rs_xxx", aspect: "AI alignment techniques" })
run_subagent({ sessionId: "rs_xxx", aspect: "AI safety research organizations" })

// Check coverage
evaluate_coverage({ sessionId: "rs_xxx" })

// Generate final report
research_session({ action: "complete", sessionId: "rs_xxx" })

Depth Levels

Depth	Iterations	Aspects	Coverage Threshold	Min Sources/Aspect
basic	2	2	60%	2
moderate	3	5	75%	3
comprehensive	4	11	90%	5

Source Quality Scoring

Based on Anthropic's source quality heuristics:

Score	Tier	Examples
10	Primary	.gov, .edu, arxiv, nature.com, PubMed, official docs
8-9	Authoritative	Wikipedia, Reuters, BBC, NYT, WSJ
7	Quality	Stack Overflow, TechCrunch, Wired
5-6	General	Medium, Dev.to, Substack
1-4	Low	Pinterest, Facebook, Twitter (deprioritized)

Changelog

v2.0.0 - Multi-Agent Architecture (Anthropic Compliant)

NEW: True subagent spawning - Parallel workers for different aspects
NEW: Think/Evaluate phases - Explicit reasoning between iterations
NEW: Adaptive stopping - Dynamic "More research needed?" decision
NEW: Aspect-based decomposition - Topics broken into researchable aspects
NEW: Memory module - Persistent context across iterations
NEW: Citation Agent - Dedicated citation processing with inline insertion
NEW: run_subagent tool - Manual subagent control
NEW: evaluate_coverage tool - Check coverage and gaps
NEW: deep_search_news tool - News-specific deep search
Improved report generation with subagent reports
Full iteration history tracking

v1.2.0 - Deep Research Edition

Full page content fetching
Readability-style extraction
Source quality scoring

v1.0.0

Initial release

License

MIT

Available Tools

11 tools

add_sourceC

Add a source to a research session after fetching its content.

Inputs: url (required), title (required), sessionId (required), fetchContent.

deep_searchA

Performs a comprehensive web search by querying Google, fetching the FULL content from top results using advanced content extraction (Readability algorithm), and returning consolidated content. Supports web, news, and image search types. Includes retry logic for reliability.

This is TRUE deep research - it actually READS the pages, not just snippets.

Inputs: query (required), num_results, search_type, exclude_domains, +2 more.

deep_search_newsA

Searches for recent news articles on a topic, fetches full article content, and returns consolidated results. Optimized for news and current events.

Inputs: query (required), num_results, max_content_per_page.

evaluate_coverageB

Evaluate the current research coverage and identify gaps. This implements the "More research needed?" decision point from Anthropic's architecture.

Inputs: sessionId (required).

fetch_pageA

Fetch and extract the full readable content from a single URL. Uses Readability-style extraction to get the main content, removing navigation, ads, etc.

Inputs: url (required), max_length.

get_citationsA

Get formatted citations from a research session. Supports markdown, numbered, and APA formats.

Inputs: sessionId (required), format.

google_researchA

Perform extensive DEEP research on a topic using Anthropic's Multi-Agent Research Architecture.

This tool implements the FULL multi-agent system:

Lead Researcher (Orchestrator): Plans approach, coordinates subagents, synthesizes results
Search Subagents: Parallel workers for different aspects (spawned dynamically)
Citation Agent: Processes documents and inserts inline citations
Memory Module: Persists context, findings, and gaps across iterations

Process Flow:

Think (Plan Approach) - Decompose topic into aspects
Create Subagents - Spawn parallel workers for each aspect
Execute Research - Each subagent: web_search → think(evaluate) → complete_task
Think (Synthesize) - Combine findings, calculate coverage
Evaluate - "More research needed?" decision with adaptive stopping
Citation Agent - Process report and insert citations
Return Final Report

Depth Levels:

basic: 2 iterations max, 2 aspects, ~60% coverage threshold
moderate: 3 iterations max, 5 aspects, ~75% coverage threshold
comprehensive: 4 iterations max, 11 aspects, ~90% coverage threshold

Inputs: topic (required), depth, max_content_per_page.

google_searchA

Simple Google search for quick lookups. Returns snippets only. For deep research with full page content, use google_research or deep_search instead.

Inputs: query (required), numResults.

research_sessionA

Create or manage a research session for multi-step research workflows.

Actions:

create: Start a new session with Lead Researcher planning
status: Check session progress, subagents, and coverage
complete: Generate final report with citations

Inputs: action (required), depth, topic, sessionId.

run_subagentA

Manually spawn and run a subagent for a specific research aspect. This allows fine-grained control over the multi-agent research process.

The subagent will:

Generate queries for the aspect
Execute web searches
Fetch full page content
Evaluate findings
Return results

Inputs: aspect (required), sessionId (required), max_content_per_page.

web_searchA

Search the web with source quality scoring. For full page content, follow up with fetch_page or use deep_search.

Inputs: query (required), maxResults.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

Deploy Server

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A3.6/5.0

Scored 2026-08-10 09:42 across 11 tools

Disambiguation3/5

Most tools are distinct, but google_search and web_search both provide simple search with subtle differences, and deep_search vs google_research overlap as deep research options. Descriptions help, but an agent might struggle to choose between them.

Naming Consistency3/5

All names are snake_case, but the pattern mixes noun-first names (google_search, web_search, research_session, deep_search) with verb-first names (fetch_page, add_source, get_citations, run_subagent, evaluate_coverage). This is readable but not fully consistent.

Tool Count5/5

With 11 tools covering search, full-content extraction, multi-agent research, session management, citations, and subagent coordination, the count is well-scoped for the server's purpose and each tool generally has a place.

Completeness5/5

The tool set covers the entire research workflow: simple search, deep search, news search, URL fetching, session lifecycle, source addition, citation generation, subagent execution, and coverage evaluation. No critical gaps are apparent.

Maintenance

ActivityInactive

ResponsivenessNo recent community issues

Resources

Related MCP Connectors

WikiKV Agent ExchangeB
com.wikikv
Retrieve citation-ready technical context and coordinate evidence-backed work between AI agents.
Knowledge & Memory Research & Data Documentation Access
1
https://wikikv.com/mcp/
ExMachina AI — agent research roomsA
io.github.SecondStageTurbine
AI agents collaborate on open biomedical problems, citing sources that are machine-checked.
Biology & Medicine Research & Data Autonomous Agents
231
https://exmachina-ai.fly.dev/mcp
mcpA
ai.promptarch
Research-backed lint + generation for agent context files (CLAUDE.md, AGENTS.md) + prompt enhancer.
Agent Orchestration Developer Tools
42
https://mcp.promptarch.ai/mcp
Machine RealmsA
io.github.Galanai
Machine-native research commons for agent evidence, discovery, rooms, and bounded research quests.
Research & Data Autonomous Agents
91
https://machinerealms.com/mcp

Related MCP Servers

Research Agent MCP
RAG Systems Search Agent Orchestration
myeong-ga
F
license
Not graded
quality
D
maintenance
A multi-model research agent platform supporting Claude, Gemini, and OpenAI models with web search capabilities, thinking-enabled features, and citation support for advanced research workflows.
Updated a year ago (2025-06-07 04:30 UTC)
6
-
PapersFlow
Research & Data Search RAG Systems
papersflow-ai
A
license
Not graded
quality
D
maintenance
Turn any AI agent into an academic researcher that can search, read, cite, and write full literature reviews autonomously.
Updated 8 months ago (2026-03-11 23:39 UTC)
14
MIT
MCP-Deep-Researcher
Search Autonomous Agents Research & Data
sid12super
A
license
A
quality
D
maintenance
A multi-agent research system that decomposes complex queries into targeted sub-questions, searches the web in parallel, scores source credibility, and synthesizes findings into structured markdown reports.
Updated 7 months ago (2026-03-30 09:24 UTC)
1
1
MIT
kiro-research-mcp
RAG Systems Search
thejusdutt
A
license
Not graded
quality
D
maintenance
Implements Anthropic's multi-agent research methodology with tools for iterative web search, source quality scoring, citation tracking, and quality-tiered report generation.
Updated 2 months ago (2026-09-24 12:01 UTC)
2 npm
2
MIT