Integration

Serpex integration

Serpex is a web search and page extraction API for AI agents. Run a web search (with site:, -site:, after: and before: operators) and get ranked results with optional page content as markdown for the top 5 or 10 results, extract up to 10 known URLs as clean markdown or HTML, and check the organization's credit balance and request counts. Searches and extractions spend prepaid credits (errors and the organization's own repeats within 5 minutes are free); the usage check is free. The GET form of search is not offered separately: it runs the same search as the POST tool. Authentication is a Serpex API key (sk_...) sent as a Bearer token; create it in the Serpex dashboard under API Keys.

Connect Serpex How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Serpex.Built from the 3 Serpex tools below, on the connection you add once.

Common Serpex use cases

giving an agent fresh web results
reading articles found by search_web
turning a page into LLM-ready text
finding a company's pages with site:

What you can do with Serpex

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Each badge shows the default agent permission of an endpoint (allowed: an agent calls it without asking; ask first: you are asked before an agent calls it, when the agent or chat asks permission for sensitive actions; never: no agent may call it), which every user can change endpoint by endpoint in their credential settings.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Usage, 1 operation

get_usageGet usageGETAgents: ask firstReturns the organization's credit balance, request counts (total, successful, failed, no results) for the last days (default 30) and its 10 most recent requests. Free: costs no credits.

Web page, 1 operation

extract_pagesExtract pagesPOSTAgents: ask firstFetches up to 10 known URLs and returns each page's main content as markdown (default) or its full HTML, with a per-URL success flag and error_type. 1 credit per URL extracted; failed URLs are free.

Web search result, 1 operation

search_webSearch webPOSTAgents: ask firstSearches the web and returns ranked results (title, url, snippet, position). include_content=true also returns the top 5 or 10 pages as markdown. 1 credit; with content up to 3 (5 pages) or 6 (10 pages), by pages delivered.

Use Serpex from Claude Code, Cursor or Codex

Connect Serpex in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Serpex operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Serpex

Create your workspace
Sign up free. Nothing to install.
Connect Serpex once
Add your Serpex connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.

Serpex integration FAQ

What can you do with the Serpex integration?

LiveContext exposes 3 operations on Serpex, covering usage, web page and web search result. Common uses include giving an agent fresh web results, reading articles found by search_web and turning a page into LLM-ready text.