Search, rerank, fetch top results, then synthesize a summary with citations using a local LLM (qwen3:14b). Returns a structured answer with source attribution. Falls back to raw fetched content if Ollama is unavailable. Best for deep research where you want pre-digested synthesis rather than raw ...
AI agents call search_and_summarize to retrieve information from Searxng without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and processes information from web search results and fetches content, then synthesizes it into a summary. These are read-only operations with no side effects on any system—no data is created, modified, deleted, or financial transactions initiated. The local LLM summarization is deterministic text processing of retrieved data.
From the tool's definition Tool performs search, fetching, and summarization operations: 'Search, rerank, fetch top results, then synthesize a summary'. All operations are retrieval and analysis based on existing data with 'source attribution' and 'Falls back to raw fetched content'.
Documented attack patterns abuse exactly the kind of access search_and_summarize gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Searxng, and nothing reaches the server without passing your rules. This is the rule we recommend for search_and_summarize:
{
"version": "1",
"default": "deny",
"tools": {
"search_and_summarize": {}
}
}search_and_summarize is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Search, rerank, fetch top results, then synthesize a summary with citations using a local LLM (qwen3:14b). Returns a structured answer with source attribution. Falls back to raw fetched content if Ollama is unavailable. Best for deep research where you want pre-digested synthesis rather than raw pages. It is categorised as a Read tool in the Searxng MCP Server, which means it retrieves data without modifying state.
Register the Searxng MCP server in PolicyLayer and add a rule for search_and_summarize: 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 Searxng. Nothing to install.
search_and_summarize 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_and_summarize 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_and_summarize. 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_and_summarize is provided by the Searxng MCP server (tadmstr/searxng-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Searxng, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
6 Searxng tools catalogued and risk-classified — across an index of 43,000+ MCP servers.