检索 MCP-Prototype 规范，ai 助手可以参考此规范来创建 html 原型页面
AI agents call getSpec to retrieve information from MCP-Prototype without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves specification documentation that an AI assistant can reference when creating HTML prototype pages. It performs a read-only operation to fetch reference material, with no capability to modify, delete, or execute code. The specification is informational only, making this a standard Read operation with low risk of misuse.
From the tool's definition Tool name is 'getSpec' and description states it 'retrieves MCP-Prototype specification' (检索 MCP-Prototype 规范). The verb 'retrieves' indicates data retrieval without side effects.
Attacks that exploit this kind of access
检索 MCP-Prototype 规范，ai 助手可以参考此规范来创建 html 原型页面. It is categorised as a Read tool in the MCP-Prototype MCP Server, which means it retrieves data without modifying state.
Register the MCP-Prototype MCP server in PolicyLayer and add a rule for getSpec: 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 MCP-Prototype. Nothing to install.
getSpec 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 getSpec 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 getSpec. 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.
getSpec is provided by the MCP-Prototype MCP server (llxxbb/mcp-prototype). 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 →