Integration

PreyReach integration

PreyReach finds local businesses (places on Google Places, not people) from a plain-English prompt such as 'dentists in Austin with no website': it searches Google Places live, checks each business's website, scores the results and returns them with phone, website and email. Agents can run a lead search and wait for its result, read a past search, enrich a finished search's leads with emails, phones and social profiles, and extract the contact details of up to 20 websites. Leads carry the contact details businesses publish or that public sources hold; for a sole trader, or an email that names an owner or employee, they identify a person and are personal data. Searches and website enrichment spend PreyReach credits. Not exposed: the routes of PreyReach's own web app on the same host (chat, saved leads, credits, onboarding, reports, admin), which sign in with a session rather than the API key, and the token register and revoke routes. Authentication is a PreyReach API key sent as a Bearer token.

Connect PreyReach How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in PreyReach.Built from the 4 PreyReach tools below, on the connection you add once.

What you can do with PreyReach

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.

Lead search, 3 operations

enrich_searchEnrich searchPOSTAgents: ask firstEnriches a completed search: tries every lead with a website not tried before for its email, phone and social profiles, saves the merged result and returns it. Not charged again (HTTP 409 if the search is not completed).
get_searchGet searchGETAgents: ask firstReturns a search made with this API key: its prompt, status (queued, running, completed or failed), timestamps, credits charged, the error if it failed, and its result (the leads) once completed.
search_leadsSearch leadsPOSTAgents: ask firstSearches local businesses from a plain-English prompt (trade, city, 'no website') and returns the leads in result. Spends credits, about one per place. Waits for the job and returns the finished result; gives up after 10 minutes.

Website, 1 operation

enrich_websitesEnrich websitesPOSTAgents: ask firstExtracts the email, phone and social profile links of up to 20 business websites. Charges one credit per website attempted.

Use PreyReach from Claude Code, Cursor or Codex

Connect PreyReach in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its PreyReach 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 PreyReach

Create your workspace
Sign up free. Nothing to install.
Connect PreyReach once
Add your PreyReach 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.