AI agents use clean_pet to create or update resources in MCPet — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your MCPet environment.
This tool modifies pet data (hygiene/cleanliness status) reversibly without deleting data or executing arbitrary code. It fits the Write category as it creates or modifies data in a reversible manner. Severity is low because the blast radius of misuse is confined to cosmetic pet state changes with no financial, destructive, or security impact.
From the tool's definition Tool name 'clean_pet' and description 'Clean your virtual pet' indicate a state modification action on a virtual pet entity. The sibling tools (feed_pet, play_with_pet, put_to_bed) confirm this is part of pet care mechanics that change pet state.
Documented attack patterns abuse exactly the kind of access clean_pet gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and MCPet, and nothing reaches the server without passing your rules. This is the rule we recommend for clean_pet:
{
"version": "1",
"default": "deny",
"tools": {
"clean_pet": {
"limits": [
{
"counter": "clean_pet_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}clean_pet stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Clean your virtual pet. It is categorised as a Write tool in the MCPet MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the MCPet MCP server in PolicyLayer and add a rule for clean_pet: 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 MCPet. Nothing to install.
clean_pet is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the clean_pet 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 clean_pet. 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.
clean_pet is provided by the MCPet MCP server (shreyaskarnik/mcpet). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from MCPet, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
6 MCPet tools catalogued and risk-classified — across an index of 43,000+ MCP servers.