We're rolling out three major improvements to our External API Service that make integrating Profound data into your workflows faster, safer, and more flexible.

Starting today, developers can query visibility reports, citation data, and bot analytics through native TypeScript and Python SDKs, plus a new Model Context Protocol (MCP) server that brings Profound directly into your AI tools.

What's New

MCP Support

The new Profound MCP server connects ChatGPT, Claude Desktop, Cursor, Cline, and other MCP-compatible tools directly to your Agent Analytics and visibility data. Query bot logs, pull citation reports, or analyze visibility trends without leaving your workflow. Your AI assistant becomes a direct interface to Profound's observability platform.

TypeScript SDK

Our TypeScript package provides full type safety across visibility reports, bot analytics, and citation data. Get autocomplete for filtering bot logs by user agent, querying visibility by domain and model, or retrieving citation breakdowns—all with compile-time guarantees. Built for teams that need reliability at scale.

Python SDK

The Python SDK gives data scientists direct access to raw bot logs and visibility metrics. Build automated alerting pipelines when crawler behavior changes, train models on citation patterns, or generate custom visibility dashboards—all in the language your team already uses.

Why It Matters

Teams tracking AI visibility shouldn't need to manually pull reports or parse API responses. These updates remove that friction.

MCP support means your AI tools can answer questions like "Which bots crawled our site in the last 5 minutes?" or "What's our visibility trend for ChatGPT this week?" directly in your chat interface. TypeScript and Python SDKs mean your engineers spend less time on API boilerplate and more time building automated monitoring systems that catch changes in real time.

For teams running Profound at scale—whether that's real-time bot alerting, daily visibility dashboards, or custom citation analysis—these updates make integration cleaner, faster, and more maintainable.

Get Started

All three packages are available now:

  • MCP Server: npx -y profound-mcp
  • TypeScript SDK: bun install profoundai
  • Python SDK: pip install profound

Full documentation is available at docs.tryprofound.com/sdks/overview, including authentication guides, example implementations, and API reference docs.

Want to see how these tools can accelerate your team's workflow? Reach out for a technical walkthrough.