Airbyte MCP and context layer beta announcement

Hey there! The Airbyte MCP and context layer are now in open beta. You can start using them right now. Run Airbyte from your AI agent The Airbyte MCP is live. Add it to Claude, ChatGPT, or any MCP-compatible agent and manage Airbyte: Create and manage sources, destinations, and connections Troubleshoot failing syncs Build and manage custom connectors Render dashboards and reports Get answers from Airbyte’s docs and source code Add the Airbyte MCP to your agent to get started. It’s free. Query your warehouse and sources from your agent The same MCP can now query your data, not just manage the pipelines that move it. Your agent writes the SQL, runs it, and brings back the answer. You don’t need a separate warehouse MCP, and you never leave the conversation. Snowflake and BigQuery: query your warehouse directly. Check whether the data landed, spot gaps or duplicates after a sync, and explore tables without opening a SQL editor. 50+ source APIs: query sources directly, without waiting for a sync. Compare a record against your warehouse, see what changed since the last run, or grab a few records on-demand. Give your team self-serve answers: Anyone with the Airbyte MCP can ask the warehouse a question in plain English. Fewer ad hoc SQL requests land on your plate, and access stays read-only. Turning it on: the Airbyte Context Layer is available to Airbyte Cloud customers. An org admin opts in. It’s off by default, and the MCP works without it. Read the context layer docs Bugs and requests To report bugs and request new features, use this form. We're excited to get your feedback. We have more features launching in the next few weeks, so stay tuned! Join the community Sign up for our Community Slack and join #airbyte-mcp-beta. © Airbyte, inc. All rights reserved. Airbyte, 100 California St., FL 11, San Francisco, CA 94111 Unsubscribe