Full-text search across the wiki.
AI agents call search_tool to retrieve information from MCP Mediawiki 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 data from the wiki without creating, modifying, deleting, or executing external operations. Full-text search is a read-only operation that returns results based on the search criteria provided. The low severity reflects minimal blast radius—misuse results in unwanted information disclosure only, not data corruption or execution of arbitrary code.
From the tool's definition Tool name 'search_tool' and description 'Full-text search across the wiki' indicate a query operation with no data modification or side effects.
Attacks that exploit this kind of access
Full-text search across the wiki. It is categorised as a Read tool in the MCP Mediawiki MCP Server, which means it retrieves data without modifying state.
Register the MCP Mediawiki MCP server in PolicyLayer and add a rule for search_tool: 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 Mediawiki. Nothing to install.
search_tool 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_tool 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_tool. 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_tool is provided by the MCP Mediawiki MCP server (mcp-mediawiki-crunchtools). 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 →