agent-reach-web-mcp
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@agent-reach-web-mcpsearch for recent articles about MCP servers"
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.
agent-reach-web-mcp
A web-search MCP server you can run with uvx, exposing the web channel of Agent Reach as MCP tools.
Search: Exa's free hosted MCP endpoint (no API key required)
Page reading: Jina Reader, with an Exa fallback
Read-only: nothing is posted or modified
Every tool call returns Agent Reach's unified result envelope (
ok/items/error.code/hint)
Tools
Tool | Description |
| Web search (Exa semantic search). Each result carries title / url / published_at / author and text highlights. |
| Read a web page as clean markdown text (Jina Reader → Exa fallback). |
Related MCP server: CrawlEyes
Run with uvx
# From a clone (recommended)
uvx --from . agent-reach-web-mcp
# Straight from the public GitHub repository (no authentication needed)
uvx --from git+https://github.com/denpadokei/agent-reach-web-mcp agent-reach-web-mcpagent-reach is pulled in automatically as a dependency installed from GitHub
(git+https://github.com/Panniantong/Agent-Reach), not from PyPI. Do not use the
same-named agent-reach package on PyPI — it is a different project.
MCP client configuration example
{
"mcpServers": {
"agent-reach-web-mcp": {
"command": "uvx",
"args": ["--from", "git+https://github.com/denpadokei/agent-reach-web-mcp", "agent-reach-web-mcp"]
}
}
}Development
uv venv
uv pip install -e .
uv run agent-reach-web-mcp # starts as a stdio MCP serverThe upstream call timeout (30 seconds by default) can be adjusted with the
AGENT_REACH_MCP_TIMEOUT environment variable, in seconds.
License
MIT (LICENSE). The upstream Agent Reach project is MIT as well.
This server cannot be deployed
Maintenance
Related MCP Connectors
Web search, scraping, RAG answers with citations, and translation as MCP tools.
Live AI-native web search with citations. One tool for every MCP client. Flat per-request pricing.
Web MCP: scrape/crawl sites, web search, brand assets, app stores, YouTube, Reddit, Hacker News.
Jina AI Reader/Search MCP — turn any URL into clean LLM-ready markdown, plus web search.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables web search using DuckDuckGo and content retrieval from URLs, returning markdown with pagination support.-
- AlicenseBqualityCmaintenanceEnables AI agents to perform web searches and extract full-text content from web pages via standard MCP tools, with fallback search and semantic reranking.22MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to perform web research over MCP: search with a real browser, fetch JS-rendered pages, download PDFs and convert them to Markdown, then search and page through large documents using bounded previews so the context window isn't flooded.BSD Zero Clause
- AlicenseNot gradedqualityBmaintenanceEnables MCP clients to run unified web searches across multiple providers such as Exa, Serper, Tavily, Firecrawl, and Jina, fusing results with reciprocal rank fusion, and to extract readable content from public web pages, PDFs, and video pages as Markdown. Provider chains are configurable, failures are isolated per provider, and responses report attempted, succeeded, failed, and skipped channels with source tracing and honest warnings.MIT