asking-with-evidence

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Asking with evidence

基于证据查询

Every watched video sits in a persistent index. Questions about it are answered from that index — text first, frames only when needed — with timestamps, a confidence score, and an honest refusal when the video does not show the answer. Never re-run a watch for a follow-up.

每一个已观看的视频都会存储在持久化索引中。关于这些视频的问题将从该索引中获取答案——优先使用文本信息，仅在必要时调用帧画面——同时提供时间戳、置信度评分；如果视频中没有相关答案，会如实拒绝回答。后续跟进问题无需重新执行观看流程。

Answer a question

回答问题

bash

watch-skill ask <video_id-or-original-url> "<question>"

Any language works; the answer comes back in the language of the question. The engine escalates on its own when unsure (dense re-sampling, zoom-crop re-OCR, stronger model) and prints a

~N tokens saved

line.

Three rules for reading the result:

Cite the timestamps it gives you; they are real evidence, not decoration.
Trust the refusal. When it says the video does not clearly show the answer, that is the answer. Do not invent one past it.
Frame paths are listed only when the engine wants you to look yourself — Read them then (or force with
```
--frames
```
).

bash

watch-skill ask <video_id-or-original-url> "<question>"

支持任意语言提问；答案会以提问语言返回。当引擎不确定时，会自动升级处理（密集重采样、缩放裁剪重新OCR、调用更强大的模型），并输出一行

~N tokens saved

的信息。

解读结果的三条规则：

引用给出的时间戳：这些是真实证据，而非装饰。
信任拒绝回答的结果：当引擎表示视频未明确展示相关答案时，这就是最终答案，不要自行编造。
仅当引擎希望你自行查看时才会列出帧路径——此时请查看这些路径（或使用
```
--frames
```
参数强制生成）。

"What happens at 2:30?"

“2:30时发生了什么？”

Moment questions get a dense window, not a whole-video ask:

bash

watch-skill moment <video_id> 2:30 [--window 10]

Returns frames + transcript + OCR around that timestamp.

针对特定时刻的问题会使用密集窗口查询，而非全视频查询：

bash

watch-skill moment <video_id> 2:30 [--window 10]

返回该时间戳附近的帧画面、转录文本及OCR识别结果。

Don't know which video? Search them all

不知道是哪个视频？搜索全部视频

bash

watch-skill search "<phrase>"

Hybrid keyword + semantic search across every video ever watched, with per-script normalization (Arabic folding, CJK segmentation, Thai segmentation). Follow a hit with

ask

moment

on that video.

bash

watch-skill search "<phrase>"

对所有已观看视频执行混合关键词+语义搜索，支持按脚本规范化处理（阿拉伯语折叠、CJK分词、泰语分词）。找到目标视频后，可对其使用

ask

或

moment

命令。

When the user corrects you

当用户纠正你的回答时

Report it so the next answer is better — see the

learning-from-mistakes

skill.

请反馈该情况，以便优化后续答案——详见

learning-from-mistakes

技能。