AI agents call natural_language_search to retrieve information from BnF API Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves or queries data from the Gallica digital library without modifying, deleting, or executing code. It produces no side effects beyond returning search results. It is analogous to the sibling search tools on the same server. The low severity reflects minimal blast radius if misused—an AI agent could only retrieve unwanted information, not harm the system or data.
From the tool's definition Tool name is 'natural_language_search' and sibling tools include 'advanced_search', 'search_by_author', 'search_by_date', 'search_by_document_type', 'search_by_subject', and 'search_by_title'.
Documented attack patterns abuse exactly the kind of access natural_language_search gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and BnF API Server, and nothing reaches the server without passing your rules. This is the rule we recommend for natural_language_search:
{
"version": "1",
"default": "deny",
"tools": {
"natural_language_search": {}
}
}natural_language_search 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.
natural_language_search. It is categorised as a Read tool in the BnF API Server MCP Server, which means it retrieves data without modifying state.
Register the BnF API Server MCP server in PolicyLayer and add a rule for natural_language_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 BnF API Server. Nothing to install.
natural_language_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 natural_language_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 natural_language_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.
natural_language_search is provided by the BnF API Server MCP server (kryzo/mcp-bibliotheque_nationale_de_france). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from BnF API Server, 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.
8 BnF API Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.