Query a RAG collection using natural language to retrieve relevant document chunks. Performs semantic search over the collection's indexed documents and returns the most relevant chunks ranked by similarity. Optionally synthesizes an AI-generated answer using the retrieved context. Parameters: - ...
Risk signalsAccepts freeform code/query input (query)
Part of the Mcp server.
Free to start. No card required.
AI agents call rag_query to retrieve information from Mcp without modifying any data. This is common in research, monitoring, and reporting workflows where the agent needs context before taking action. Because read operations don't change state, they are generally safe to allow without restrictions -- but you may still want rate limits to control API costs.
Even though rag_query only reads data, uncontrolled read access can leak sensitive information or rack up API costs. An agent caught in a retry loop could make thousands of calls per minute. A rate limit gives you a safety net without blocking legitimate use.
Read-only tools are safe to allow by default. No rate limit needed unless you want to control costs.
{
"version": "1",
"default": "deny",
"tools": {
"rag_query": {}
}
}See the full Mcp policy for all 47 tools.
These attack patterns abuse exactly the kind of access rag_query gives an agent. Each links to the full case and the policy that stops it:
Other read tools across the catalogue. The same approach applies to each: allow, with a rate cap to control cost.
Query a RAG collection using natural language to retrieve relevant document chunks. Performs semantic search over the collection's indexed documents and returns the most relevant chunks ranked by similarity. Optionally synthesizes an AI-generated answer using the retrieved context. Parameters: - query: Natural language question or search phrase - top_k: Number of chunks to retrieve (default 5, max 20) - threshold: Minimum similarity score 0-1 (only return chunks above this score) - synthesize: If true, uses an LLM to generate a natural language answer from the retrieved chunks (default false — returns raw chunks only) - model: LLM model to use for synthesis (only relevant when synthesize is true, default: anthropic/claude-haiku-4.5) - filter: Metadata filter to narrow results (e.g. { category: "faq" }) Example — raw retrieval: Input: { app_id: "app_abc123", collection: "knowledge-base", query: "How do I reset my password?", top_k: 3 } Output: { chunks: [ { text: "To reset your password, go to Settings > Security > Reset Password...", score: 0.92, document_id: "doc_abc", metadata: { category: "faq", source: "help-center" } }, ... ] } Example — with synthesis: Input: { app_id: "app_abc123", collection: "knowledge-base", query: "How do I reset my password?", top_k: 5, synthesize: true } Output: { answer: "To reset your password, navigate to Settings > Security and click...", chunks: [ ... ], model: "gpt-4o-mini" } Example — with metadata filter: Input: { app_id: "app_abc123", collection: "knowledge-base", query: "pricing plans", filter: { category: "billing", version: "2.0" } } Use this to: - Search documentation or knowledge bases using natural language - Build AI-powered Q&A features for end users - Find relevant context for AI assistants - Power search bars with semantic understanding Common errors: - RESOURCE_NOT_FOUND: App or collection doesn't exist - COLLECTION_EMPTY: No documents have been ingested yet Idempotency: Safe to call anytime (read-only operation).. It is categorised as a Read tool in the Mcp MCP Server, which means it retrieves data without modifying state.
Register the MCP server in PolicyLayer and add a rule for rag_query: 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. Nothing to install.
rag_query 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 rag_query 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 rag_query. 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.
rag_query is provided by the MCP server (@butterbase/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 47 Mcp tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4,600+ MCP servers and 31,000+ tools scanned and risk-classified.