Ask a follow-up question in an existing Genie conversation and return SQL + query result.
AI agents invoke follow_up to trigger actions in databricks-genie-MCP. 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.
The tool executes SQL queries against Databricks, generating and running SQL as part of the follow-up interaction. While it may appear conversational, it triggers actual SQL execution whose effects depend on the natural language input. Since the SQL is generated dynamically and executed against a Databricks environment, this is Execute-level risk.
From the tool's definition Ask a follow-up question in an existing Genie conversation and return SQL + query result
Documented attack patterns abuse exactly the kind of access follow_up gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and databricks-genie-MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for follow_up:
{
"version": "1",
"default": "deny",
"tools": {
"follow_up": {
"limits": [
{
"counter": "follow_up_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}follow_up stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Ask a follow-up question in an existing Genie conversation and return SQL + query result. It is categorised as a Execute tool in the databricks-genie-MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the databricks-genie- MCP server in PolicyLayer and add a rule for follow_up: 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 databricks-genie-MCP. Nothing to install.
follow_up 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 follow_up 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 follow_up. 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.
follow_up is provided by the databricks-genie- MCP server (yashshingvi/databricks-genie-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from databricks-genie-MCP, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
3 databricks-genie-MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.