AI agents use unsubscribe-from-thread to create or update resources in Liveblocks — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Liveblocks environment.
Unsubscribing from a thread modifies the user's subscription state (removes a subscription record), which is a reversible write operation — the user can re-subscribe at any time. It does not delete data, execute code, or involve financial transactions. Blast radius is low as it only affects notification preferences for a single thread.
From the tool's definition unsubscribe-from-thread / Unsubscribe from a Liveblocks thread
Documented attack patterns abuse exactly the kind of access unsubscribe-from-thread gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Liveblocks, and nothing reaches the server without passing your rules. This is the rule we recommend for unsubscribe-from-thread:
{
"version": "1",
"default": "deny",
"tools": {
"unsubscribe-from-thread": {
"limits": [
{
"counter": "unsubscribe-from-thread_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}unsubscribe-from-thread 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.
Unsubscribe from a Liveblocks thread. It is categorised as a Write tool in the Liveblocks MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Liveblocks MCP server in PolicyLayer and add a rule for unsubscribe-from-thread: 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 Liveblocks. Nothing to install.
unsubscribe-from-thread 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 unsubscribe-from-thread 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 unsubscribe-from-thread. 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.
unsubscribe-from-thread is provided by the Liveblocks MCP server (liveblocks/liveblocks-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Liveblocks, 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.
39 Liveblocks tools catalogued and risk-classified — across an index of 43,000+ MCP servers.