Web search. Returns titles, URLs, snippets.
AI agents call searxng_web_search to retrieve information from Mcp Searxng without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs a web search query and returns read-only information (search results). There are no side effects, no data modification, no code execution, and no irreversible actions. It is purely a retrieval mechanism that queries a search engine and returns results to the user. This is a classic Read category operation with minimal risk from misuse.
From the tool's definition Tool description states it 'Returns titles, URLs, snippets' with no mention of modification, deletion, execution, or financial operations. The name 'web_search' and the return of search results (titles, URLs, snippets) indicate data retrieval only.
Attacks that exploit this kind of access
Web search. Returns titles, URLs, snippets. It is categorised as a Read tool in the Mcp Searxng MCP Server, which means it retrieves data without modifying state.
Register the Mcp Searxng MCP server in PolicyLayer and add a rule for searxng_web_search: 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 Searxng. Nothing to install.
searxng_web_search 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 searxng_web_search 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 searxng_web_search. 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.
searxng_web_search is provided by the Mcp Searxng MCP server (mcp-searxng). 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 →