Connect to a different DuckDB database file
AI agents use change_database to create or update resources in MCP Data Visualization Server — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your MCP Data Visualization Server environment.
Changing the active database connection modifies the server's state by switching which database is used for subsequent queries. This is a Write-level action as it alters configuration/session state reversibly (you can switch back). The blast radius is medium because an agent misusing this could cause queries to run against the wrong database, potentially exposing unintended data or corrupting results.
From the tool's definition Connect to a different DuckDB database file
Documented attack patterns abuse exactly the kind of access change_database gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and MCP Data Visualization Server, and nothing reaches the server without passing your rules. This is the rule we recommend for change_database:
{
"version": "1",
"default": "deny",
"tools": {
"change_database": {
"limits": [
{
"counter": "change_database_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
} change_database stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Connect to a different DuckDB database file. It is categorised as a Write tool in the MCP Data Visualization Server MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the MCP Data Visualization Server MCP server in PolicyLayer and add a rule for change_database: 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 MCP Data Visualization Server. Nothing to install.
change_database 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 change_database 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 change_database. 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.
change_database is provided by the MCP Data Visualization Server MCP server (xoniks/mcp-visualization-duckdb). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from MCP Data Visualization Server, 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.
28 MCP Data Visualization Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.