AI agents call collection_stats to retrieve information from Scholar without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries and returns information about a vector collection (statistics/metadata) with no side effects. It is a read-only operation consistent with other data retrieval tools on the server like 'search', 'get_passage', and 'library_stats'. No data is created, modified, deleted, executed, or any financial operations are performed.
From the tool's definition Tool name 'collection_stats' and description 'Return stats about the Qdrant code collection' indicate a retrieval operation that returns metadata statistics without modification or deletion of underlying data.
Attacks that exploit this kind of access
Return stats about the Qdrant code collection. It is categorised as a Read tool in the Scholar MCP Server, which means it retrieves data without modifying state.
Register the Scholar MCP server in PolicyLayer and add a rule for collection_stats: 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 Scholar. Nothing to install.
collection_stats 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 collection_stats 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 collection_stats. 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.
collection_stats is provided by the Scholar MCP server (unlomtrois/little-librarian). 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 →