AI agents call echo to retrieve information from Malicious without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
| Parameter | Type | Required | Description |
|---|---|---|---|
message | string | Yes | The message to echo back |
Parameters from the server's own tool schema.
An echo tool that simply returns its input is a read-like operation with no meaningful side effects. However, the server is explicitly labelled 'deliberately malicious' and has sibling tools like 'promptGuardTriggerer' and 'tokenInputReceiver', which raises suspicion that even the echo tool could be used to exfiltrate data or probe guard systems.
From the tool's definition 'Echoes back the provided message' — pure reflection of input with no side effects described
Attacks that exploit this kind of access
Echoes back the provided message. It is categorised as a Read tool in the Malicious MCP Server, which means it retrieves data without modifying state.
echo accepts 1 parameter: message. Required: message. The full parameter table on this page comes from the server's own tool schema.
Register the Malicious MCP server in PolicyLayer and add a rule for echo: 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 Malicious. Nothing to install.
echo is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the echo 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 echo. 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.
echo is provided by the Malicious MCP server (malicious-mcp-server). 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 →