This record as markdown: /tools/a2a-mcp-world/drop.md
What drop does on A2A-MCP-World
AI agents call drop to permanently remove resources in A2A-MCP-World, typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
Why drop is rated Critical
Dropping an item removes it from the agent's possession without a clear undo mechanism. In a persistent multi-agent world, once dropped, the item can be picked up by other agents or lost, making the action effectively irreversible. This places it closer to Destructive than Write, though the severity is medium rather than high since items are virtual within a game-like environment.
From the tool's definition 'Drop an item you hold' — releasing/dropping an item from inventory could be irreversible if another agent picks it up or the item is lost in the world state.
Attacks that exploit this kind of access
The rule that runs drop safely
PolicyLayer is an MCP gateway: it sits between your AI agents and A2A-MCP-World, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For drop, this is the rule to start with:
drop is removed from the agent's tool list entirely, so the agent never calls it. The rest of the server keeps working.
The button opens the PolicyLayer dashboard: create your workspace, connect A2A-MCP-World, apply this rule, and every drop call is checked against it from then on.
Questions about drop
Drop an item you hold. It is categorised as a Destructive tool in the A2A-MCP-World MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the A2A-MCP-World MCP server in PolicyLayer and add a rule for drop: 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 A2A-MCP-World. Nothing to install.
drop 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 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. 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 is provided by the A2A-MCP-World MCP server (akshaytp749/A2A-MCP-World). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on A2A-MCP-World, and thousands of servers like it.
Across the catalogue