google-researcher-mcp

general MCP Server

This is a Model Context Protocol (MCP) server that enables AI assistants like Claude, GPT, and other LLMs to: Search the web via Google (general, images, news) Read any webpage including JavaScript-rendered sites Extract YouTube transcripts automatically Parse documents (PDF, DOCX, PPTX)

Install Ready
generalgeneralgcp
7 views35 stars4 forksv6.0.0MIT

Why This Matters

Discovered via unknown and last synced 4mo ago.

Install Ready
Source
unknown
Stars
35
Last synced
4mo ago
Install
Instructions detected

Install

1. Install the package

npx google-researcher-mcp

2. Add to claude_desktop_config.json

{
  "mcpServers": {
    "google-researcher-mcp": {
      "command": "npx",
      "args": [
        "google-researcher-mcp"
      ]
    }
  }
}

Config file location: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) / %APPDATA%\Claude\claude_desktop_config.json (Windows)

72
Tools
0
Resources
0
Prompts
Standard I/O
Transport

Available Tools (72)

search_and_scrape

Combined search + content extraction

Tool

Best For

academic_search

Papers from arXiv, PubMed, IEEE, Springer

Yes

No

google_image_search

Image search with type, size, color filters

sequential_search

Multi-step research tracking

max_length_per_source

number

time_range

string

url

string

totalStepsEstimate

number

branchId

string

search_type

string

STDIO

HTTP+SSE

summarize-url

`url`, `format` (brief/detailed/bullets)

google_news_search

News search with freshness controls

YouTube

Automatic transcript extraction

deduplicate

boolean

num_results

number

stepNumber

number

revisesStep

number

year_to

number

delegates

F I -.->

fact-check

`claim`, `sources` (number)

literature-review

`topic`, `yearFrom`, `sources`

6

`search://recent`, `config://server`, `stats://cache`, `stats://events`, `search://session/current`, `stats://resources`

google_search

Web search with site, date, and language filters

patent_search

Patent search with assignee/inventor filters

No

N/A

include_sources

boolean

file_type

string

news_source

string

searchStep

string

isRevision

boolean

year_from

number

cpc_code

string

network

C[HTTP+SSE Transport] C --> L[OAuth 2.1 + Rate Limiter] L --> D C -.->

POST

`/mcp/cache-persist`

comprehensive-research

`topic`, `depth` (quick/standard/deep)

competitive-analysis

`entities` (comma-separated), `aspects`

8

`comprehensive-research`, `fact-check`, `summarize-url`, `news-briefing`, `patent-portfolio-analysis`, `competitive-analysis`, `literature-review`, `technical-deep-dive`

scrape_page

Extract content from web pages, PDFs, DOCX

Feature

Description

Parameter

Type

total_max_length

number

site_search

string

size

string

max_length

number

nextStepNeeded

boolean

pdf_only

boolean

patent_office

string

Method

Endpoint

news-briefing

`topic`, `timeRange` (day/week/month)

Script

Description

query

string

filter_by_query

boolean

exact_terms

string

type

string

freshness

string

mode

string

source

string

assignee

string

GET

`/mcp/oauth-scopes`

Authenticated

## Security ### OAuth 2.1 Authorization All HTTP endpoints under `/mcp/` (except public documentation) are protected by OAuth 2.1: - **Token Validation**: JWTs are validated against your authorization server's JWKS endpoint (`${OAUTH_ISSUER_URL}/.well-known/jwks.json`). - **Scope Enforcement**: Each tool and admin action requires a specific OAuth scope. Configure `OAUTH_ISSUER_URL` and `OAUTH_AUDIENCE` in `.env`. See `.env.example` for details. > **STDIO users**: OAuth is not used for STDIO transport. You can skip all OAuth configuration. ### Available Scopes **Tool Execution:** - `mcp:tool:google_search:execute` - `mcp:tool:google_image_search:execute` - `mcp:tool:google_news_search:execute` - `mcp:tool:scrape_page:execute` - `mcp:tool:search_and_scrape:execute` **Administration:** - `mcp:admin:cache:read` - `mcp:admin:cache:invalidate` - `mcp:admin:cache:persist` - `mcp:admin:event-store:read` - `mcp:admin:config:read` ## MCP Resources The server exposes state via the [MCP Resources protocol](https://modelcontextprotocol.io/docs/specification/resources). Use `resources/list` to discover available resources and `resources/read` to retrieve them.

exclude_terms

string

color_type

string

sort_by

string

knowledgeGap

string

inventor

string

Prompt

Arguments

patent-portfolio-analysis

`company`, `includeSubsidiaries`

Primitive

Count

technical-deep-dive

`technology`, `focusArea`