AI agents call search_similar to retrieve information from Knowledge Rag without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Despite empty description, the name and server context strongly indicate this is a read-only search/retrieval operation with no side effects. It queries the knowledge base to find similar items, consistent with standard RAG functionality. No modification, deletion, or external execution capability is implied. Confidence lowered slightly (0.85 vs 0.95) due to missing explicit description.
From the tool's definition Tool name 'search_similar' in a RAG (Retrieval-Augmented Generation) system context implies similarity-based search/retrieval. Sibling tools include 'search_knowledge' and retrieval-focused operations.
Documented attack patterns abuse exactly the kind of access search_similar gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Knowledge Rag, and nothing reaches the server without passing your rules. This is the rule we recommend for search_similar:
{
"version": "1",
"default": "deny",
"tools": {
"search_similar": {}
}
}search_similar 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_similar. It is categorised as a Read tool in the Knowledge Rag MCP Server, which means it retrieves data without modifying state.
Register the Knowledge Rag MCP server in PolicyLayer and add a rule for search_similar: 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 Knowledge Rag. Nothing to install.
search_similar 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_similar 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_similar. 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_similar is provided by the Knowledge Rag MCP server (lyonzin/knowledge-rag). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Knowledge Rag, 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.
12 Knowledge Rag tools catalogued and risk-classified — across an index of 43,000+ MCP servers.