Run a federated query across multiple connections (different dialects). Returns a unified result set with per-source provenance. [ARCHITECT tier]
AI agents invoke cross_db_query to trigger actions in ThinAir Data. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
Although the tool returns a result set (suggesting a read-like outcome), it actively *executes* queries across multiple database connections and dialects. The blast radius is high because: (1) the query content is arbitrary and not constrained to SELECT — it could include DML or DDL depending on the database permissions; (2) it spans multiple connections simultaneously, multiplying potential impact; (3) 'federated'…
From the tool's definition "Run a federated query across multiple connections (different dialects)" — the word 'Run' indicates active execution of queries, not merely reading schema or metadata.
Attacks that exploit this kind of access
Run a federated query across multiple connections (different dialects). Returns a unified result set with per-source provenance. [ARCHITECT tier]. It is categorised as a Execute tool in the ThinAir Data MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the ThinAir Data MCP server in PolicyLayer and add a rule for cross_db_query: 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 ThinAir Data. Nothing to install.
cross_db_query is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the cross_db_query 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 cross_db_query. 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.
cross_db_query is provided by the ThinAir Data MCP server (thinairtelematics/thinair-data). 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 →