Send a message to display in the chat as the assistant. Supports markdown formatting.
AI agents use sendMessage to create or update resources in Mup — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Mup environment.
This tool creates new data (chat messages) that is visible and persistent within the chat history (evidenced by sibling tools 'getHistory', 'clearHistory', 'loadSession'). While reversible (messages can be cleared via clearHistory), it modifies the conversational state and record. This is a Write operation rather than Read (retrieves existing data) or Execute (triggers external side effects).
From the tool's definition Tool name 'sendMessage' with description 'Send a message to display in the chat as the assistant' directly indicates creation/modification of chat message data. The tool writes/creates new message content in the chat interface.
Documented attack patterns abuse exactly the kind of access sendMessage gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Mup, and nothing reaches the server without passing your rules. This is the rule we recommend for sendMessage:
{
"version": "1",
"default": "deny",
"tools": {
"sendMessage": {
"limits": [
{
"counter": "sendmessage_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}sendMessage 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.
Send a message to display in the chat as the assistant. Supports markdown formatting. It is categorised as a Write tool in the Mup MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Mup MCP server in PolicyLayer and add a rule for sendMessage: 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 Mup. Nothing to install.
sendMessage 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 sendMessage 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 sendMessage. 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.
sendMessage is provided by the Mup MCP server (mup-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Mup, 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.
6 Mup tools catalogued and risk-classified — across an index of 43,000+ MCP servers.