AI agents call drop_ai_model to permanently remove resources in Mcp Oceanbase — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
The word 'drop' is a definitive destructive action in database and machine learning domains. Dropping an AI model removes it irreversibly and cannot be undone. While the empty description reduces certainty slightly, the unambiguous semantics of the verb 'drop' place this in the Destructive category rather than lower severity categories.
From the tool's definition Tool name 'drop_ai_model' uses the verb 'drop', which in database and ML contexts means irreversible deletion or removal of an entity. The empty description limits confirmation but the semantic meaning of 'drop' strongly indicates data destruction.
Documented attack patterns abuse exactly the kind of access drop_ai_model gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Mcp Oceanbase, and nothing reaches the server without passing your rules. This is the rule we recommend for drop_ai_model:
{
"version": "1",
"default": "deny",
"hide": [
"drop_ai_model"
]
}drop_ai_model disappears from the agent's tool list entirely, and any attempt to call it is denied. The rest of the server keeps working.
Free to start. No card required.
drop_ai_model. It is categorised as a Destructive tool in the Mcp Oceanbase MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Mcp Oceanbase MCP server in PolicyLayer and add a rule for drop_ai_model: 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 Mcp Oceanbase. Nothing to install.
drop_ai_model 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 drop_ai_model 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 drop_ai_model. 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.
drop_ai_model is provided by the Mcp Oceanbase MCP server (oceanbase/awesome-oceanbase-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 134 Mcp Oceanbase tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
134 Mcp Oceanbase tools catalogued and risk-classified — across an index of 42,500+ MCP servers.