Fable's judgement

One of the most interesting tips I got from the Fireside Chat I hosted with Cat Wu and Thariq Shihipar from the Claude Code team at AIE on Wednesday was to let Fable (and to a certain extent Opus) use their own judgement rather than dictating how they should work.
周三我在AIE主持与Cat Wu和Thariq Shihipar（来自Claude Code团队）的炉边谈话时，获得的一个最有趣建议是：让Fable（以及一定程度上Opus）自主判断工作方式，而非硬性规定它们该如何工作。

The example they gave was testing. You can tell Fable “only use automated testing for larger features, don’t update and run tests for small copy or design changes” - but it’s better to just tell Fable to use its own judgement when deciding to write tests instead.
他们给的例子是测试。你可以告诉Fable“只在大型功能上使用自动化测试，不要为小的文案或设计变更更新和运行测试”——但更好的做法是直接让Fable自行判断何时编写测试。

Jesse Vincent just gave me a related tip to help avoid burning too many of those valuable Fable tokens in the few days we have left before the prices go up. Tell Fable to use other models for smaller tasks, applying its own judgement about which model to use.
杰西·文森特刚刚给了我一个相关建议，帮助避免在我们为数不多的几天涨价前消耗太多宝贵的Fable令牌。告诉Fable在小任务上使用其他模型，让它自行判断使用哪个模型。

I prompted Claude Code just now with:
我刚才用以下内容提示了Claude Code：

For all coding tasks use your judgement to decide an appropriate lower power model and run that in a subagent

Claude saved this memory file in ~/.claude/projects/name-of-project/memory/delegate-coding-to-subagents.md:
Claude将这个记忆文件保存在~/.claude/projects/name-of-project/memory/delegate-coding-to-subagents.md：

---
name: delegate-coding-to-subagents
description: Simon wants coding tasks delegated to subagents running an appropriately lower-power model
metadata: 
  node_type: memory
  type: feedback
  originSessionId: 30068d78-43a9-4fb1-bb29-9799e18c526a
---
Stated by Simon on 2026-07-03: “For all coding tasks use your judgement to decide an appropriate lower power model and run that in a subagent.”
Simon 于 2026-07-03 指出：“对于所有编程任务，请自行判断并选择一个合适的低功耗模型，在子代理中运行它。”

Why: cost/efficiency — implementation work rarely needs the top-tier model; judgment, review, and synthesis stay with the main loop.
原因：成本/效率——实现类工作很少需要顶级模型；判断、审查和综合工作仍由主循环负责。

How to apply: when a task in this project is primarily writing/editing code, spawn an Agent with a model override (sonnet for substantive implementation, haiku for trivial/mechanical edits) and a self-contained prompt; review the result in the main loop before committing. Design, auditing, data synthesis, and anything judgment-heavy stays in the main model. See also [[project-goals]].
应用方式：当此项目中的任务主要是编写/编辑代码时，需生成一个带有模型覆盖的代理（实质性实现用 Sonnet，机械性简单编辑用 Haiku）和自包含提示；在主循环中审查结果后再提交。设计、审计、数据合成以及任何需要判断力的工作仍由主模型负责。另见 [[项目目标]]。

So far it seems to be working well. I’m getting a ton of work done and my Fable allowance is shrinking less quickly than before.
迄今为止效果不错。我完成了大量工作，同时 Fable 额度消耗速度也比之前慢了。