AI agents call zurich_sparql to retrieve information from Zurich Open Data MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
SPARQL is a query language for RDF data that retrieves information without modifying underlying data. This classifies as Read. Severity is high because: (1) the tool provides access to 900+ datasets including potentially sensitive data (parliamentary proceedings, possibly location data, environmental monitoring); (2) an AI agent could broadly query and aggregate sensitive information across multiple datasets; (3)…
From the tool's definition Tool named 'zurich_sparql' with empty description; inferred from context as a SPARQL query interface to 900+ open datasets including environmental, mobility, geodata, and parliamentary data. SPARQL enables querying and retrieving structured data.
Documented attack patterns abuse exactly the kind of access zurich_sparql gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Zurich Open Data MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for zurich_sparql:
{
"version": "1",
"default": "deny",
"tools": {
"zurich_sparql": {}
}
}zurich_sparql is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
zurich_sparql. It is categorised as a Read tool in the Zurich Open Data MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Zurich Open Data MCP Server MCP server in PolicyLayer and add a rule for zurich_sparql: 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 Zurich Open Data MCP Server. Nothing to install.
zurich_sparql is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the zurich_sparql 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 zurich_sparql. 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.
zurich_sparql is provided by the Zurich Open Data MCP Server MCP server (pypi:zurich-opendata-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Zurich Open Data MCP 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.
24 Zurich Open Data MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.