Delete a document by ID.
AI agents call delete_document to permanently remove resources in Rag — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
This tool permanently removes data (documents) from the knowledge store without possibility of reversal. This is a destructive operation that cannot be undone. While the blast radius is limited to individual documents rather than entire databases, the irreversible nature of deletion and potential loss of ingested knowledge across sessions justifies the Destructive category and high severity.
From the tool's definition Tool name is 'delete_document' and description states 'Delete a document by ID.' The verb 'delete' combined with the action of removing a document by its identifier indicates irreversible data deletion.
Attacks that exploit this kind of access
Delete a document by ID. It is categorised as a Destructive tool in the Rag MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Rag MCP server in PolicyLayer and add a rule for delete_document: 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 Rag. Nothing to install.
delete_document is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the delete_document 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 delete_document. 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.
delete_document is provided by the Rag MCP server (mrankitvish/rag-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 →