Token-optimized document search using LODA (LLM-Optimized Document Access). Returns relevant sections within optional token budget. Uses Bloom filters for O(1) section elimination and caching for 10x+ speedup.
AI agents call loda_search to retrieve information from LODA MCP Server 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 document sections without any side effects, modification of data, code execution, or destructive operations. The optimization techniques (Bloom filters, caching) are implementation details that don't change the fundamental nature of the operation. The low severity reflects minimal blast radius - misuse would only return document content the AI agent could already access directly.
From the tool's definition Tool description explicitly states it 'Returns relevant sections' and performs 'document search' - a retrieval operation with no modification or execution capabilities.
Attacks that exploit this kind of access
Token-optimized document search using LODA (LLM-Optimized Document Access). Returns relevant sections within optional token budget. Uses Bloom filters for O(1) section elimination and caching for 10x+ speedup. It is categorised as a Read tool in the LODA MCP Server MCP Server, which means it retrieves data without modifying state.
Register the LODA MCP Server MCP server in PolicyLayer and add a rule for loda_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 LODA MCP Server. Nothing to install.
loda_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 loda_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 loda_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.
loda_search is provided by the LODA MCP Server MCP server (patrickkarle/loda-mcp-server). 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 →