collection_schema
AI agents call collection_schema to retrieve information from Legacy Mongodb without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool appears to retrieve schema metadata from MongoDB collections without modifying data. The server explicitly guarantees read-only access, and the tool name indicates schema inspection rather than modification. Low severity because schema information exposure has minimal immediate blast radius, though confidence is moderately reduced due to empty tool description.
From the tool's definition Server is described as 'read-only access to legacy MongoDB instances' and tool name 'collection_schema' suggests retrieval of schema information.
Attacks that exploit this kind of access
collection_schema. It is categorised as a Read tool in the Legacy Mongodb MCP Server, which means it retrieves data without modifying state.
Register the Legacy Mongodb MCP server in PolicyLayer and add a rule for collection_schema: 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 Legacy Mongodb. Nothing to install.
collection_schema 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_schema 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_schema. 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_schema is provided by the Legacy Mongodb MCP server (webxspark/legacy-mongodb-mcp). 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 →