What AI agents actually do with your MCP server
说明与用途
MCP 服务器看似健康,但代理会遇到隐藏的失败,例如缺失工具、错误参数、描述不清、循环调用以及大响应导致的令牌消耗增加。
问题
MCP servers appear healthy but agents encounter hidden failures such as missing tools, bad arguments, misleading descriptions, looping calls, and large responses that inflate token usage.
结果
By measuring agent interactions, the team uncovered hidden failure modes that were not visible in standard logs, leading to concrete fixes that improved agent reliability and reduced token costs.
工作流
- Instrument server with mcpspan
- Run simulated traffic
- Collect tool call data
- Analyze missing tool calls
- Identify argument validation failures
- Adjust tool descriptions
- Detect and fix looping calls
- Measure response sizes
- Implement limits or summaries
证据
Agents call tools you don't have
Bad arguments are the most common failure, and they hide
The tool description is code
Agents loop, and every single call looks fine
Response size is invisible on latency charts
- 记录日期: 2026-10-07T10:55:59.930Z