Skip to main content
Glama
slothitude

Web Eyes

by slothitude
OverviewSchemaRelated ServersScoreDiscussions
Python
Hybrid

Web Eyes

Search, crawl, and summarize the web — exposed as both a REST API and an MCP server for LLM agents.

Powered by SearXNG, Crawl4AI, and NVIDIA NIM.

What it does

Web Eyes provides a pipeline of web intelligence tools:

  1. Search — query SearXNG for web results

  2. Crawl — extract clean text from URLs using a headless browser

  3. Summarize — distill content via an LLM (NVIDIA NIM)

  4. Ask — full pipeline: search → crawl → synthesize an answer with citations

  5. See — take screenshots and use a vision LLM to extract content from JS-heavy, canvas-rendered, or image-heavy pages

  6. Look — analyze any image directly via vision AI (no URL crawling needed)

These are exposed via a FastAPI REST API and an MCP (Model Context Protocol) server, so any MCP-compatible agent (Claude Desktop, Claude Code, Cursor, etc.) can use them directly.

Related MCP server: websearch-mcp

Quick Start

1. Start SearXNG

docker compose up -d

2. Configure environment

cp .env.example .env
# Edit .env and set NIM_API_KEY (get one at https://build.nvidia.com/)

3. Install dependencies

pip install -r requirements.txt

4. Run

REST API + MCP together (port 3000):

python main.py

  • REST API: http://localhost:3000

  • MCP endpoint: http://localhost:3000/mcp

  • Interactive docs: http://localhost:3000/docs

Standalone MCP server:

python run_mcp.py              # stdio (default)
python run_mcp.py http         # streamable-http on port 3001
python run_mcp.py sse          # SSE on port 3001

REST API

Method

Path

Description

POST

/search

Search → crawl → summarize

POST

/crawl

Crawl specific URLs

POST

/summarize

Crawl + summarize specific URLs

POST

/ask

Search → crawl → answer with citations

POST

/see

Screenshot + vision extraction + summarize

POST

/look

Analyze a base64-encoded image with vision AI

Example:

curl -X POST http://localhost:3000/search \
  -H "Content-Type: application/json" \
  -d '{"query": "latest Rust release", "limit": 5}'

MCP Tools

Tool

Parameters

Description

search_web

query, limit=10

Search, crawl, and summarize

crawl_pages

urls

Extract raw text from URLs

summarize_pages

urls, instruction?

Crawl and summarize URLs

ask_web

question, scrape_top=3

Answer a question with web sources

see_pages

urls, instruction?, extract_prompt?

Screenshot + vision extraction + summarize

look_at_image

image_base64, instruction?

Analyze an image directly with vision AI

Agent Configuration

Claude Desktop / Claude Code (mcp.json):

{
  "mcpServers": {
    "web-eyes": {
      "command": "python",
      "args": ["C:\\Users\\you\\web_eyes\\run_mcp.py", "stdio"]
    }
  }
}

Remote agents (HTTP transport):

http://localhost:3001/mcp

Configuration

All settings are in .env. See .env.example for defaults.

Variable

Default

Description

NIM_API_KEY

—

NVIDIA NIM API key (required for summarize/ask)

NIM_BASE_URL

https://integrate.api.nvidia.com/v1

NIM API endpoint

NIM_MODEL

google/gemma-3-27b-it

LLM model for summarization

NIM_VISION_MODEL

google/gemma-3-27b-it

Vision model for screenshot extraction

VISION_FALLBACK_ENABLED

true

Auto-fallback to vision when text extraction fails

VISION_WORD_THRESHOLD

30

Minimum words before triggering vision fallback

VISION_MAX_IMAGE_DIMENSION

1280

Max screenshot dimension before resize

SEARXNG_HOST

localhost

SearXNG host

SEARXNG_PORT

8888

SearXNG port

API_HOST

0.0.0.0

REST API bind address

API_PORT

3000

REST API port

MCP_HOST

0.0.0.0

Standalone MCP bind address

MCP_PORT

3001

Standalone MCP port

Project Structure

web_eyes/
├── main.py           FastAPI app (REST + mounted MCP)
├── mcp_server.py     MCP server with 6 tools
├── run_mcp.py        Standalone MCP entry point
├── controller.py     Core pipeline logic
├── search.py         SearXNG search client
├── crawler.py        Crawl4AI web crawler
├── summarizer.py     NIM LLM summarization + vision extraction
├── vision.py         Image resize and message utilities
├── config.py         Environment config
├── logger.py         Rich logging
├── docker-compose.yml
├── requirements.txt
└── searxng/
    └── settings.yml  SearXNG configuration

License

MIT

This server cannot be installed

F
license - not found
-
quality - not tested
D
maintenance

How are these scores calculated?

Maintenance

–Maintainers
–Response time
–Release cycle
–Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

  • Crawl4AI RAG MCP Server

    Chillbruhhh
    A
    license
    -
    quality
    D
    maintenance
    Provides AI agents and coding assistants with advanced web crawling and RAG capabilities, allowing them to scrape websites and leverage that knowledge through various retrieval strategies.
    Last updated
    2
    MIT

  • websearch-mcp

    roberthamel
    A
    license
    A
    quality
    -
    maintenance
    Enables web searching via SearXNG, page content extraction with Crawl4AI, and image analysis using vision language models. It provides AI agents with tools for information synthesis and web-based data retrieval through OpenAI-compatible LLM endpoints.
    Last updated
    3

View all related MCP servers

Related MCP Connectors

  • mcp

    Reliable web access for AI agents: smart HTTP, rotating proxies, and full-browser rendering.

  • Parallel Search MCP

    The best web search for your AI Agent

  • Tako

    Give your agent web search and authoritative datasets: S&P Global, FRED, OECD, SimilarWeb & more.

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/slothitude/web_eyes'

If you have feedback or need assistance with the MCP directory API, please join our Discord server