bak_get_opendata
AI agents call bak_get_opendata to retrieve information from Pypi:swiss Culture without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The naming pattern 'get_*' and context of a cultural data server strongly indicate this is a retrieval operation (Read). However, the empty description and ambiguous 'opendata' parameter reduce confidence. No evidence of write, destructive, execute, or financial capabilities. Severity is low because cultural heritage data access poses minimal risk even if misused.
From the tool's definition Tool name 'bak_get_opendata' suggests retrieval of open data; server description indicates read-only access to Swiss cultural heritage data with no API key required. Tool description is empty, limiting specificity.
Attacks that exploit this kind of access
bak_get_opendata. It is categorised as a Read tool in the Pypi:swiss Culture MCP Server, which means it retrieves data without modifying state.
Register the Pypi:swiss Culture MCP server in PolicyLayer and add a rule for bak_get_opendata: 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 Pypi:swiss Culture. Nothing to install.
bak_get_opendata 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 bak_get_opendata 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 bak_get_opendata. 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.
bak_get_opendata is provided by the Pypi:swiss Culture MCP server (malkreide/swiss-culture-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 →