AI agents use airbyte_create_source to create or update resources in Airbyte — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Airbyte environment.
The tool creates or modifies data configurations reversibly. While the description is empty (reducing confidence slightly), the function name and server context clearly indicate this tool creates new source configurations in Airbyte. This is categorized as Write rather than Execute because it configures data pipelines rather than triggering code execution.
From the tool's definition Tool name 'airbyte_create_source' indicates creation of a source object within Airbyte. Server description confirms the tool operates on 'sources' which are connection endpoints for data extraction.
Attacks that exploit this kind of access
airbyte_create_source. It is categorised as a Write tool in the Airbyte MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Airbyte MCP server in PolicyLayer and add a rule for airbyte_create_source: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Airbyte. Nothing to install.
airbyte_create_source is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the airbyte_create_source rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for airbyte_create_source. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
airbyte_create_source is provided by the Airbyte MCP server (trustxai/airbyte-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →