Delete specific observations from entities in the knowledge graph.
AI agents call delete_observations to permanently remove resources in Rawthink — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
This tool permanently removes observations (data points) from the knowledge graph, which cannot be undone. While not financial in nature and not executing arbitrary code, it is clearly destructive because it irreversibly deletes stored information. The high severity reflects that an agent misusing this could permanently lose important context or memories stored in the user's knowledge graph.
From the tool's definition Tool name is 'delete_observations' with description 'Delete specific observations from entities in the knowledge graph.' The verb 'delete' indicates irreversible removal of data.
Attacks that exploit this kind of access
Delete specific observations from entities in the knowledge graph. It is categorised as a Destructive tool in the Rawthink MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Rawthink MCP server in PolicyLayer and add a rule for delete_observations: 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 Rawthink. Nothing to install.
delete_observations 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_observations 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_observations. 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_observations is provided by the Rawthink MCP server (ygtalp/rawthink-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 →