AI agents call list_museums to retrieve information from Confd without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and queries museum data through various filters without creating, modifying, deleting, or executing any side effects. It is a standard data retrieval operation typical of search/list endpoints, placing it firmly in the Read category with low severity due to absence of any destructive or system-altering capabilities.
From the tool's definition Tool name 'list_museums' and description explicitly states it 'Search and filter museums' and 'Returns paginated results.' The operations described (search, filter, retrieve) are all read-only with no modification or deletion.
Attacks that exploit this kind of access
Search and filter museums. Supports name search, country code, source identifier, tag, published status, and discarded status filters. Returns paginated results. It is categorised as a Read tool in the Confd MCP Server, which means it retrieves data without modifying state.
Register the Confd MCP server in PolicyLayer and add a rule for list_museums: 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 Confd. Nothing to install.
list_museums 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 list_museums 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 list_museums. 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.
list_museums is provided by the Confd MCP server (mytours/confd-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 →