A simple greeting tool that returns personalized messages to users
AI agents call say_hello to retrieve information from Dev Tool MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool simply returns greeting messages—a pure read operation with no capability to modify, execute, delete, or move data. Despite being on a server with web crawling and automation capabilities, this specific tool is isolated and performs only a simple retrieval/output function. Severity is low because misuse poses no practical risk to systems or data.
From the tool's definition Tool name: 'say_hello'; description states it 'returns personalized messages to users' with no side effects or external operations mentioned.
Attacks that exploit this kind of access
A simple greeting tool that returns personalized messages to users. It is categorised as a Read tool in the Dev Tool MCP MCP Server, which means it retrieves data without modifying state.
Register the Dev Tool MCP server in PolicyLayer and add a rule for say_hello: 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 Dev Tool MCP. Nothing to install.
say_hello 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 say_hello 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 say_hello. 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.
say_hello is provided by the Dev Tool MCP server (osins/dev-tool-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 →