查看目标 agent 上 arthas 的命令帮助（权威、与实际版本一致）。
AI agents call help to retrieve information from Arthas Mcp Proxy without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool retrieves and displays help information about Arthas commands. This is a safe, informational operation with no side effects—no code execution, data modification, or destructive actions. It merely queries and returns documentation from the target Java process diagnostic tool.
From the tool's definition Tool name 'help' with description indicating it 'views command help for arthas on the target agent' (查看目标 agent 上 arthas 的命令帮助). This is a read-only operation that retrieves documentation/help text without executing commands or modifying state.
Attacks that exploit this kind of access
查看目标 agent 上 arthas 的命令帮助（权威、与实际版本一致）。. It is categorised as a Read tool in the Arthas Mcp Proxy MCP Server, which means it retrieves data without modifying state.
Register the Arthas Mcp Proxy MCP server in PolicyLayer and add a rule for help: 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 Arthas Mcp Proxy. Nothing to install.
help 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 help 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 help. 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.
help is provided by the Arthas Mcp Proxy MCP server (zhaoziqian/arthas-mcp-proxy). 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 →