Recce | Blog

🤖 Recce is now AI-Agent ready (MCP) + BigQuery Support

Written by Recce | Nov 30, 2025 8:00:00 AM

We just shipped the infrastructure that allows AI agents to interact directly with your data validation workflow. Alongside native BigQuery support and cloud-persisted checklists, this release is designed to eliminate manual friction from your review process.

🤖 Recce is now AI-Agent Ready

The biggest friction in data validation is the manual effort involved in clicking, checking, and verifying. We are changing that.

Recce now supports the Model Context Protocol (MCP). This means you can connect AI agents (like Claude Desktop or other MCP clients) directly to your Recce instance. Instead of manually clicking through lineage and diffs, you can now ask your AI agent to do the groundwork for you:

  • "Check the lineage impact of my latest PR."
  • "Run a value diff on the orders model."
  • "Summarize the downstream breakage risks."

We believe in a human-in-the-loop workflow. By giving AI agents the tools to understand your data infrastructure, we remove the busy work. This lets you focus on the high-value decisions that actually matter.

Want to see it in action?

We are hosting a deep-dive webinar to show you exactly how to set up an AI Agent to accelerate your PR reviews.

 

☁️ Cloud-Persisted Checklists

Validation is a team sport, and your "work in progress" shouldn't disappear when you close a tab. We’ve upgraded Recce Cloud so that your Checklists are now persisted in the cloud automatically.

  • Real-time Sync: See your teammate's checks appear as they run them.
  • Never Lose Work: Close your browser, switch devices, or refresh because your validation state is saved.
  • Audit Trail: Clearly see who approved which check and when.

🚀 BigQuery Support is Officially Live

You asked (a lot), and we listened. Recce Cloud now supports BigQuery natively. This is a full implementation allowing BigQuery teams to run the same powerful diffs, lineage checks, and PR validations that our Snowflake and Postgres users love. It is available immediately in General Availability.

🎙️ We Launched a Podcast: Data Renegades

We are excited to announce the launch of Data Renegades hosted by Recce’s own CL Kao and Dori Wilson. Check us out on Apple Podcasts (soon Spotify as well)

 

Ep #3: Building Tools That Shape Data with Maxime Beauchemin: The creator of Apache Airflow and Superset joins us to discuss the evolution of open source and his forward-looking perspective on agentic workflows.

Ep #2: Data Journalism Unleashed with Simon Willison The co-creator of Django and creator of Datasette dives into how LLM-based agents will redefine data cleaning, enrichment, and analysis. 

Ep #1: Introducing Data Renegades CL and Dori set the stage, discussing their background in civic tech and how we are designing software around LLMs today. 

🛠 Other Notable Updates

  • recce-cloud CLI Package: We broke this out into a lightweight package. It installs faster and is cleaner for your CI/CD pipelines.
  • Value Diffs in Context: You can now trigger a Value Diff directly from column context menus which saves you clicks when investigating data anomalies.
  • Copilot Instructions: We added copilot-instructions.md to our repo. This makes it easier for GitHub Copilot to understand your Recce configuration context.

💬 Help Shape What’s Next at Recce

With the launch of MCP and BigQuery, we are planning our roadmap for Q1. We want to hear what you need next.

  • Are you ready to use AI agents in your workflow?
  • What’s the one friction point we haven't solved yet?

 

If you can spare 30 minutes to chat, we’ll send you a $25 gift card as a thank you.

Join Our Community

Your feedback drives everything we build. Connect with us:

 
View full post