- Accueil
- Plugins
- Outils et capacités
- dsh-bilibili_video-study_buddy (plugin)
dsh-bilibili_video-study_buddy (plugin)
aaarookie/dsh-bilibili_video-study_buddy/plugin
Bilibili course study buddy for DeepSeek Harness. Four tools: study_frame grabs a timestamped frame at 1080p with region cropping and a disk cache; study_scan samples a time range into one grid image for locating; study_subs extracts AI subtitles and scores three signals to detect contaminated subtitle tracks; study_asr transcribes a short window with surrounding context. Read-only; all data stays local.
Installer
dsh plugin --profile web add github:aaarookie/dsh-bilibili_video-study_buddyCe plugin se trouve dans le sous-répertoire plugin du dépôt.
README
dsh-bilibili_video-study_buddy
Let the model look at the right frame at the right timestamp, then turn a lecture into checkable notes.
A DeepSeek Harness plugin, plus the methodology and real note artifacts behind it.
Repo: https://github.com/aaarookie/dsh-bilibili_video-study_buddy | Plugin: plugin/ (npm: dsh-bilibili_video-study_buddy) | 中文说明: README.zh.md
What this is
| Path | Contents |
|---|---|
plugin/ | The DSH bundle. Four tools: study_frame / study_scan / study_subs / study_asr |
docs/ | Methodology: ecosystem survey, plugin comparison, workflow, environment, verification log |
examples/ | Real artifacts: Karpathy GPT lecture notes (md + docx), ASR transcripts, frame grids |
scripts/ | Standalone PowerShell tools (optional, Windows) |
Three counter-intuitive findings
1. Do not feed the whole video — feed the right frames
A 30-minute lecture carries its signal in roughly 20 frames. Grab one frame at a timestamp, crop away the browser chrome, and let the model read it. That is one to two orders of magnitude cheaper than "video understanding".
2. A grid is for locating, not reading
| Approach | Size | Est. tokens |
|---|---|---|
| 6 full frames at 1080p | — | ~5,850 |
| 6-cell grid (400 px/cell) | 1224×450 | ~585 (−90%) |
At 400 px per cell you cannot read code. The grid only answers "which stretch is worth a close look".
3. Bilibili AI subtitles lie — verify before trusting
On a lecture about GPT, the subtitle endpoint returned:
[0.0s] 核能一种很硬核的能源 (nuclear energy is a hardcore energy source)
[2.4s] 比如说一克浓缩铀 (say, one gram of enriched uranium)
and on another attempt, English song lyrics. Both had nothing to do with the video.
study_subs therefore scores three independent signals — coverage, title-term hits, speech rate —
and refuses to emit the body when the verdict is FAIL.
Install
dsh plugin --profile web add dsh-bilibili_video-study_buddy
Or in DSH: Settings → Plugins → Add plugin and enter the package name. Restart DSH once.
Requirements: Node ≥ 20, ffmpeg (required); a Bilibili session (optional, unlocks 1080p and AI subtitles);
faster-whisper (only for study_asr).
Usage
You: explain the RL Training part at P8 of BV1mqrTBvEaf
The model will, on its own:
study_substo probe subtitles → verdict: contaminated (this video really is)study_scan -from 14:00 -to 18:00→ a grid that locates the slide at 16:20study_frame -at 16:20 -crop "0,0,0,190"→ one high-res framestudy_asr -at 16:20→ the lecturer's own words with surrounding context- read the frame against the transcript → explain, then write timestamped notes → Word
Measured numbers
| Item | Result |
|---|---|
| Cropped 1080p frame | 1234×800, ~975 tokens (−45% vs full frame) |
| 6-cell grid | 1224×450, ~585 tokens (−93% vs reading each) |
| 90-second ASR window | 8.8 s (faster-whisper small, CPU) |
| Full 42-minute transcript | 229 s |
| Frame cache hit | < 1 s (first fetch 13 s) |
| Understanding one 30-minute lecture | ≈ ¥0.03 (DeepSeek off-peak) |
Cost model: examples/karpathy-gpt/p8-cache/cost-analysis.md (Chinese).
Limitations (all from real failures)
- A timestamp belongs to one part, not to the whole video. Multi-part courses need the right
page. - Anonymous caps at 640×480; 1080p requires a Bilibili session.
- ASR systematically mishears code terms (
free(p)→ "4AP", 栈上 → 站上). Code always comes from the frame. - The grid cannot be used to read code — locating only.
- Read-only: never likes, comments, or coins.
Docs
English: docs/en/ | 中文: docs/zh/
| # | Topic | EN | 中文 |
|---|---|---|---|
| 01 | Survey of 102 Bilibili-related DSH plugins | en | zh |
| 02 | Nine candidates compared, three technical routes | en | zh |
| 03 | The five-step study loop and prompt templates | en | zh |
| 04 | Install, login, and error lookup | en | zh |
| 05 | Every measurement, including what was not tested | en | zh |
| 06 | The timestamp → frame → notes pipeline | en | zh |
| 07 | What was installed and how to undo it | en | zh |
| 08 | Plugin install and integration record | en | zh |
License
MIT
Plugins associés
archify (deepseek-harness)
tt-a1i/archify
WeKnora (dsh-weknora)
tencent/weknora
next-ai-draw-io (dsh-plugin)
dayuanjiang/next-ai-draw-io
weknora
tencent/weknora