VidkNot

VidkNot turns video links into structured knowledge notes. It downloads audio, transcribes speech, generates Markdown notes, and saves them to Obsidian, Feishu, Notion, Yuque, or AnythingLLM.

| English | 中文 |

• Convert courses, interviews, podcasts, and industry videos into searchable notes
• Save useful short-video content into a personal knowledge base
• Expose video-to-note capability to agents through MCP
• Reduce transcription mistakes with dual-ASR cross-validation

VidkNot is managed through the uv ecosystem. If you don't have uv yet, install it first (see the uv docs):

# macOS / Linux
curl -LsSf https://astral.sh/uv/install.sh | sh
# Windows
winget install astral-sh.uv
# or via pip
pip install uv

The current GitHub release is v0.3.2. Install this repository version from GitHub:

uv tool install "vidknot @ git+https://github.com/suonian/vidknot.git@v0.3.2"

For Feishu support:

uv tool install "vidknot[feishu] @ git+https://github.com/suonian/vidknot.git@v0.3.2"

Run it without installing:

uvx "vidknot @ git+https://github.com/suonian/vidknot.git@v0.3.2" --help

For development:

git clone https://github.com/suonian/vidknot.git
cd vidknot
uv sync

This creates a .venv from uv.lock with the dev tools (pytest, ruff) included. Run commands with uv run or activate the environment:

uv run vidknot "URL" --destinations none

FFmpeg must be available locally:

ffmpeg -version

Copy .env.example to .env and configure the keys you need:

SILICONFLOW_API_KEY=your_siliconflow_api_key
OPENAI_API_KEY=your_openai_compatible_api_key

# Optional: Feishu
FEISHU_APP_ID=your_feishu_app_id
FEISHU_APP_SECRET=your_feishu_app_secret
FEISHU_FOLDER_TOKEN=your_feishu_folder_token

# Optional: Obsidian
OBSIDIAN_VAULT_PATH=/path/to/obsidian/vault

# Optional: Notion
NOTION_TOKEN=your_notion_token
NOTION_PAGE_ID=your_notion_page_id

# Optional: Yuque
YUQUE_TOKEN=your_yuque_token
YUQUE_LOGIN=your_yuque_login

# Optional: Douyin cookie file
VIDKNOT_DOUYIN_COOKIE_FILE=/path/to/douyin-cookies.txt

Default settings live in config.yaml. Dual-ASR correction is enabled by default:

settings:
  enable_correction: true
  correction_version: v4
faster_whisper:
  model: small
  device: cpu
  compute_type: int8

v4 is the conservative default. v3 makes broader corrections and carries a higher risk of changing valid text.

CLI:

# Generate a note and save to the default destination, Obsidian
vidknot "https://v.douyin.com/example/"

# Print output only
vidknot "https://v.douyin.com/example/" --destinations none

# Save to Feishu
vidknot "https://v.douyin.com/example/" --destinations feishu

# Save to multiple destinations, comma-separated
vidknot "https://v.douyin.com/example/" --destinations obsidian,anythingllm

# Disable dual-ASR correction
vidknot "https://v.douyin.com/example/" --no-correct

# Check local requirements
vidknot --check-env

In a development checkout, prefix these with uv run (e.g. uv run vidknot --check-env).

MCP:

vidknot --mcp

FastAPI:

uv run uvicorn vidknot.api:app --reload

Python API:

from vidknot import VideoKnowledgePipeline

pipeline = VideoKnowledgePipeline(destinations="none")
result = pipeline.run("https://v.douyin.com/example/")

print(result["markdown"])

VidkNot writes Markdown notes with:

• Video title, source URL, and processing metadata
• Topic and summary
• Structured key points and details
• Important quotes
• Terms and explanations
• Timestamped transcript

• Do not commit .env, cookie files, or API keys
• Only process video content you are allowed to access and use
• Follow the terms of video platforms, cloud services, and note platforms
• Third-party service availability, pricing, and permissions are controlled by their providers

MIT