learning-from-mistakes

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Learning from mistakes

从错误中学习

When a video answer turns out wrong, the correction is worth more than an apology. Report it and Watch Skill learns locally — the mistake is classified, stored as a lesson, injected into future similar questions, and where possible the original question is immediately re-asked to confirm the lesson actually fixes it.

当视频回答出现错误时，纠正内容比道歉更有价值。上报错误后，Watch Skill会在本地学习——错误会被分类、作为经验教训存储，并注入到未来的同类问题中；在可能的情况下，会立即重新询问原问题，以确认该经验教训确实解决了问题。

Report a correction

上报纠正内容

The moment the user corrects a video answer, run:

bash

watch-skill lessons add <video_id> "<question>" "<the wrong answer>" "<the correction>"

The output includes the lesson (

error_class

guidance

) and, when the question was re-asked, whether the corrected answer now comes back — tell the user which it was. Nothing is uploaded; lessons live in

~/.watch-skill/

当用户纠正视频回答的那一刻，运行以下命令：

bash

watch-skill lessons add <video_id> "<question>" "<the wrong answer>" "<the correction>"

输出内容包含经验教训（

error_class

、

guidance

），以及当原问题被重新询问时，纠正后的回答是否返回——要告知用户结果。所有内容均不上传；经验教训存储在

~/.watch-skill/

目录下。

See what has been learned

查看已学习内容

bash

watch-skill lessons list

bash

watch-skill lessons list

The savings meter

节省统计仪表

bash

watch-skill stats

Lifetime token economics: what text-first answering and the semantic cache have saved versus stuffing raw frames into context. Quote real numbers from it when the user asks whether any of this is worth it.

bash

watch-skill stats

终身令牌经济统计：对比直接将原始帧放入上下文的方式，纯文本优先回答和语义缓存所节省的资源。当用户询问这些功能是否有价值时，可以引用其中的真实数据。