Pure MCP · no SDK, no client library

Give your AI the live web.

SearchMesh is one MCP endpoint that lets any AI search the web across many engines at once and read the full contents of any page. Paste one URL into Claude, ChatGPT, Cursor or your own agent — then just tell it to use SearchMesh.

2 tools: search + fetch Any MCP client Zero scraping code 100 concurrent, 0 dropped

Connect

Add one URL. That's the whole integration.

SearchMesh speaks the Model Context Protocol over streamable HTTP, so it plugs into anything that speaks MCP. There is nothing to install and no library to import.

# one command, then restart your session
claude mcp add --transport http searchmesh \
  https://api.searchmesh.io/mcp \
  --header "x-api-key: YOUR_KEY"

Endpoint https://api.searchmesh.io/mcp · transport streamable HTTP · auth x-api-key header.

Then just ask for it by name

Your AI sees SearchMesh in its toolbox and reaches for it on its own — but you can also aim it directly:

you →Use SearchMesh to find what changed in the latest Postgres release.
you →Blast SearchMesh for reviews of this laptop, then read the top three.
you →Search the web and fetch that page so you're not guessing.

Tools

Two things your model cannot do alone.

Search finds the pages. Fetch reads them. Together they turn a model that guesses into one that checks.

web_search

Search the live web

Queries a mesh of independent engines in parallel and returns ranked results with titles, URLs and snippets. Filter by recency, language or result count. Built for the questions a model would otherwise answer from stale memory — prices, versions, releases, people, news.

read_url

Fetch page content

Give it any URL and get back clean, readable markdown — HTML converted, JSON pretty-printed, boilerplate stripped. Use it on results from a search, or on any link a user drops into the conversation.

search_suggestions

Sharpen a vague query

Autocomplete-style alternatives, so an agent can refine a fuzzy or misspelled question before spending a search on it.

mesh_status

See the mesh

Reports which upstream engines and categories are live right now. Diagnostic — handy when you want to know what the mesh is drawing from.

Why a mesh

One search engine is a single point of failure.

Most search APIs resell one upstream source. When it rate-limits you, you get nothing back. SearchMesh is built the other way around.

Many engines, one answer

Every query fans out to several independent search engines at once and the results are merged and ranked. If one engine throttles, the others still answer.

Multi-IP egress

Rate limits are enforced per source IP, so requests leave over more than one, and each engine is routed through the path it actually accepts.

A queue, not a rejection

Burst past capacity and requests wait their turn instead of failing. Measured at 100 simultaneous searches: every one answered, none dropped.

Pricing

Pay for requests. Nothing else.

One request is one web_search or one read_url. Suggestions and status checks are free. Unused requests do not roll over.

Free

Wire it up and see if your agent likes it.
$0
250 requests / month
  • Both tools, no limits on features
  • Full result depth
  • Community support
Start free

Starter

A personal agent that searches all day.
$9
5,000 requests / month
  • 5,000 requests / month
  • Recency & language filters
  • Email support
Choose Starter
MOST POPULAR

Pro

Production agents and small teams.
$29
25,000 requests / month
  • 25,000 requests / month
  • Higher burst concurrency
  • Usage dashboard
  • Priority support
Choose Pro

Scale

Heavy pipelines and research swarms.
$79
100,000 requests / month
  • 100,000 requests / month
  • Highest burst concurrency
  • Multiple API keys
  • Priority support
Choose Scale

What a request costs you

PlanRequests / monthMonthlyYearlyPer request
Free250$0$0
Starter5,000$9$90$0.0018
Pro25,000$29$290$0.0012
Scale100,000$79$790$0.0008
EnterprisecustomTalk to usvolume

Yearly plans are billed as ten months — two months free. Plans renew automatically and can be cancelled any time; you keep the requests you have paid for until the period ends.

FAQ

Questions worth asking first.

Which AIs can use this?

Anything that speaks MCP. That includes Claude (Code, Desktop and the web app's custom connectors), OpenAI's Agents SDK, Cursor, Windsurf, Zed, LangChain's MCP adapters, and any agent you write yourself. If your client cannot do MCP, the endpoint is still plain JSON-RPC over HTTPS and you can call it with curl.

Do I need to write any scraping code?

No. That is the entire point. You add one URL and a key; SearchMesh handles the engines, the rate limits, the retries and the HTML-to-markdown conversion.

How do I make my AI actually use it?

Usually you don't have to — the tools describe themselves clearly enough that a model reaches for them when a question needs current information. When you want to be explicit, say "use SearchMesh" or "blast SearchMesh" and it will.

What counts as a request?

One web_search call or one read_url call. A search that returns thirty results is still one request. search_suggestions and mesh_status are not billed.

What happens if I go over my plan?

Calls are paused rather than silently billed, and we tell you before you get there. You can upgrade mid-cycle and only pay the difference.

Is billing live yet?

Not yet — payments are being wired up now. The plans above are the pricing we will launch with. Until then keys are issued manually, so get in touch and we'll set you up.