展示HN：ToolGuard – 用于AI代理工具调用的Pytest
我厌倦了我的AI代理因为大型语言模型（LLM）错误地生成JSON键或传递字符串而不是整数而崩溃。因此，我构建了ToolGuard——它通过边界情况（如空值、缺失字段、类型不匹配、10MB负载）对你的Python工具函数进行模糊测试，并给出一个可靠性评分，满分为100%。<p>运行测试不需要LLM。它读取你的类型提示，生成Pydantic模式，并以确定性的方式破坏功能。<p>使用以下命令安装：pip install py-toolguard<p>GitHub: <a href="https://github.com/Harshit-J004/toolguard" rel="nofollow">https://github.com/Harshit-J004/toolguard</a><p>如果你正在构建复杂的工具链，我将非常荣幸你能查看这个仓库。非常欢迎对架构提出严厉的反馈！
查看原文
I got tired of my AI agents crashing because the LLM hallucinated a JSON key or passed a string instead of an int. So I built ToolGuard — it fuzzes your Python tool functions with edge-cases (nulls, missing fields, type mismatches, 10MB payloads) and gives you a reliability score out of 100%.<p>No LLM needed to run tests. It reads your type hints, generates a Pydantic schema, and deterministically breaks things.<p>pip install py-toolguard<p>GitHub: <a href="https://github.com/Harshit-J004/toolguard" rel="nofollow">https://github.com/Harshit-J004/toolguard</a><p>If you are building complex tool chains, I would be incredibly honored if you checked out the repo. Brutal feedback on the architecture is highly encouraged!