Save, remove, list, or check dataset search watches for monitoring new datasets
AI agents use watch_query to create or update resources in MobusMCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your MobusMCP environment.
The tool manages persistent watch/subscription configurations: it can save (create) and remove (delete) watches. However, 'remove' of a watch is a reversible configuration change (not irreversible data destruction), and 'list/check' are read operations. The dominant action is Write (creating/modifying saved watches).
From the tool's definition Save, remove, list, or check dataset search watches for monitoring new datasets
Documented attack patterns abuse exactly the kind of access watch_query gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and MobusMCP, and nothing reaches the server without passing your rules. This is the rule we recommend for watch_query:
{
"version": "1",
"default": "deny",
"tools": {
"watch_query": {
"limits": [
{
"counter": "watch_query_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}watch_query 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.
Save, remove, list, or check dataset search watches for monitoring new datasets. It is categorised as a Write tool in the MobusMCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Mobus MCP server in PolicyLayer and add a rule for watch_query: 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 MobusMCP. Nothing to install.
watch_query 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 watch_query 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 watch_query. 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.
watch_query is provided by the Mobus MCP server (mobus-ai/mobus). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from MobusMCP, 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.
13 MobusMCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.