AI agents call search_users to retrieve information from Kwork without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Searching users is fundamentally a retrieval operation that queries user information on the Kwork marketplace. Even if it returns user profiles, contact info, or other metadata, this is read-only access with no ability to modify or delete data. The low confidence reflects the missing description, which would have provided definitive confirmation of the tool's scope and any potential filtering/privacy implications.
From the tool's definition Tool name 'search_users' indicates a query/lookup operation. The empty description prevents confirmation of the exact behavior, but search operations typically retrieve data without side effects.
Attacks that exploit this kind of access
search_users. It is categorised as a Read tool in the Kwork MCP Server, which means it retrieves data without modifying state.
Register the Kwork MCP server in PolicyLayer and add a rule for search_users: 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 Kwork. Nothing to install.
search_users 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 search_users 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 search_users. 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.
search_users is provided by the Kwork MCP server (simonether/kwork-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 →