S1 S1

AI 是一个写代码极快的实习生——但它不懂什么是工程。OPC 的故事从这个矛盾开始。

AI is a blazingly fast intern — but it doesn't understand what engineering means. The OPC story begins with this contradiction.

4 种角色、14 个节点、1686 行机械规范——OPC 把「让 AI 帮我看看」变成了一条有状态的流水线。

4 roles, 14 nodes, 1686 lines of mechanical spec — OPC turns 'let AI take a look' into a stateful pipeline.

安全审计 47 分，命令注入、路径穿越、空文件作弊——三周硬化之后，OPC 学会了一件事：不让 AI 变好，让坏后果变小。

Security audit score: 47. Command injection, path traversal, empty file cheating — after three weeks of hardening, OPC learned one thing: don't make AI better, make bad outcomes smaller.

从 hardcoded 到 capability contract——OPC 的扩展系统设计，以及为什么「关节是扩展点，骨头中间不能挂肌肉」。

From hardcoded to capability contracts — OPC's extension system design, and why 'joints are extension points, you can't hang muscles in the middle of a bone.'

我跟 AI 说「跑完再来找我」，然后去睡觉了。几个小时后回来，一句话就让分数从 0.487 跳到了 0.68——不是因为 AI 做得好。

I told AI 'finish up and come find me,' then went to bed. Hours later I checked back — one sentence pushed the score from 0.487 to 0.68. Not because AI did great work.

23 小时、$92、44 个 AI 员工——从零到一个完整产品的真实账单。以及为什么 Loop 执行计划，不生成计划。

23 hours, $92, 44 AI employees — the real bill from zero to a complete product. And why loops execute plans, they don't generate them.

用 OPC 审计 OPC，所有 gate 都 PASS——这才是最大的问题。一个关于镜子、洋葱和盲点的故事。

Using OPC to audit OPC, every gate PASS — that's the real problem. A story about mirrors, onions, and blind spots.

跑到第 8 个 tick，AI 开始重复工作。它忘了自己刚修过这个 bug。因为它的记忆被压缩掉了。

By tick 8, AI started repeating work. It forgot it had just fixed this bug. Because its memory got compressed away.

OPC 所有设计决策背后的同一个原则：不要优化 AI 的能力，要优化它出错时的后果。Gate 是底线不是来源，AI 能做相对判断不能做绝对评价。

One principle behind every OPC design decision: don't optimize AI's capability, optimize the consequences when it fails. Gates are floors, not sources. AI can make relative judgments, not absolute ones.

AI 跑了 23 小时方向全错，一句话就扳回来了。autonomous loop 不是不需要人，而是需要人在对的时间说对的话。

The AI ran 23 hours in the wrong direction. One sentence fixed it. Autonomous loops don't eliminate humans—they redefine when humans matter.

Gate 全绿但封面图缩了 40%。review 系统 review 自己时崩了。当安全网本身有洞，修复不是回滚——是让系统学到教训。S1 终章。

Gates all green, but the cover image shrunk 40%. The review system crashed when reviewing itself. When safety nets have holes, recovery isn't rollback—it's making the system learn. S1 finale.