Leave a project gracefully. Archives your unread messages and removes your membership. You can rejoin later to reclaim your agent name. Session must be initialized by calling status tool first.
AI agents use leave_project to create or update resources in Brainstorm — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Brainstorm environment.
This tool modifies state by removing the agent's membership from a project and archiving messages. It is reversible since the description explicitly states 'You can rejoin later to reclaim your agent name,' making it Write rather than Destructive. Misuse could disrupt coordination workflows, hence medium severity.
From the tool's definition Leave a project gracefully. Archives your unread messages and removes your membership.
Documented attack patterns abuse exactly the kind of access leave_project gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Brainstorm, and nothing reaches the server without passing your rules. This is the rule we recommend for leave_project:
{
"version": "1",
"default": "deny",
"tools": {
"leave_project": {
"limits": [
{
"counter": "leave_project_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
} leave_project 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.
Leave a project gracefully. Archives your unread messages and removes your membership. You can rejoin later to reclaim your agent name. Session must be initialized by calling status tool first. It is categorised as a Write tool in the Brainstorm MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Brainstorm MCP server in PolicyLayer and add a rule for leave_project: 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 Brainstorm. Nothing to install.
leave_project 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 leave_project 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 leave_project. 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.
leave_project is provided by the Brainstorm MCP server (theodorstorm/brainstorm-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Brainstorm, 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.
17 Brainstorm tools catalogued and risk-classified — across an index of 43,000+ MCP servers.