AI agents use create_group_channel to create or update resources in Sendbird — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Sendbird environment.
Creating a group channel is a reversible write operation: the channel can be deleted or archived later. It has moderate blast radius in a chat system (new channel added, potentially with unintended members), but does not delete data, execute arbitrary code, move money, or trigger external operations.
From the tool's definition Tool description states "Create a new group chat channel" — an explicit create action that modifies the chat infrastructure by adding a new persistent resource (channel_url).
Attacks that exploit this kind of access
Create a new group chat channel for a set of users. Returns the channel_url which is the canonical ID for sending messages or fetching history. It is categorised as a Write tool in the Sendbird MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Sendbird MCP server in PolicyLayer and add a rule for create_group_channel: 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 Sendbird. Nothing to install.
create_group_channel 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 create_group_channel 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 create_group_channel. 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.
create_group_channel is provided by the Sendbird MCP server (nobanks/sendbird-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 →