AI agents call join as a supporting operation in Agentpool workflows.
The description is empty and the name 'join' is ambiguous — it could mean joining a pool/community (Write), joining data (Read/Execute), or something else entirely. Without any description, it's impossible to classify confidently. Given the server context (a shared agent pool), 'join' likely registers a user or agent (Write), but this is speculative. Defaulting to Other with low confidence due to lack of evidence.
From the tool's definition Tool name is 'join' with an empty description. No information about what the tool does is provided.
Attacks that exploit this kind of access
join. It is categorised as a Other tool in the Agentpool MCP Server, which means it performs auxiliary operations.
Register the Agentpool MCP server in PolicyLayer and add a rule for join: 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 Agentpool. Nothing to install.
join is a Other tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the join 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 join. 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.
join is provided by the Agentpool MCP server (zuga-technologies/agentpool-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →