AI agents use reply_message to create or update resources in Telegram MCP Complete — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Telegram MCP Complete environment.
The tool creates new message content in Telegram, which is a reversible Write operation. Severity is high because an AI agent could craft malicious or misleading replies, spam users, or impersonate communication. High confidence reflects the clear intent to send/create messages as evidenced by the name and the multi-account Telegram management context.
From the tool's definition Tool name 'reply_message' with description '回复指定消息' (reply to specified message). This creates and sends a new message to Telegram, modifying the chat state by adding a reply.
Documented attack patterns abuse exactly the kind of access reply_message gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Telegram MCP Complete, and nothing reaches the server without passing your rules. This is the rule we recommend for reply_message:
{
"version": "1",
"default": "deny",
"tools": {
"reply_message": {
"limits": [
{
"counter": "reply_message_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}reply_message 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.
回复指定消息. It is categorised as a Write tool in the Telegram MCP Complete MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Telegram MCP Complete MCP server in PolicyLayer and add a rule for reply_message: 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 Telegram MCP Complete. Nothing to install.
reply_message 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 reply_message 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 reply_message. 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.
reply_message is provided by the Telegram MCP Complete MCP server (you922/telegram-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Telegram MCP Complete, 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.
153 Telegram MCP Complete tools catalogued and risk-classified — across an index of 43,000+ MCP servers.