byted-tos-video-process
Inspects videos, extracts frames, and concatenates video clips stored in Volcengine TOS: read video metadata, capture single snapshots, batch-capture multiple frames by explicit timestamps or interval, concatenate multiple video clips, and save results locally or back to TOS. Use this skill when the user needs to get video duration or codec info, capture a poster image or thumbnail, extract frames at specific timestamps, sample frames at regular intervals, merge or join multiple video clips into one — even if they don't explicitly mention 'video processing' or 'frame extraction'.
What this skill does
# Volcengine TOS Video Process
Inspect videos, extract frames, and concatenate video clips stored in Volcengine TOS — metadata lookup, single snapshot, multi-frame capture, video concatenation, and TOS-to-TOS persistence.
## Setup (once per environment)
Install dependencies on first use:
```bash
cd {baseDir}
pip install -r {baseDir}/requirements.txt
```
Then run scripts with Python 3.7+:
```bash
python3 {baseDir}/scripts/<script>.py <args>
```
If you see a `ModuleNotFoundError` for `tos`, reinstall dependencies.
## Environment Variables
This skill relies on the TOS identity declared in the `metadata` block. Common runtime variables are:
| Environment Variable | Required | Description |
| --- | --- | --- |
| `TOS_ACCESS_KEY` | Yes | TOS access key ID |
| `TOS_SECRET_KEY` | Yes | TOS secret access key |
| `TOS_ENDPOINT` | Yes | TOS endpoint URL |
| `TOS_REGION` | Yes | TOS region |
| `TOS_BUCKET` | Yes | Source bucket that stores the video |
| `TOS_OBJECT_KEY` | No | Source object key of the video. Some scripts accept alternatives, while `video_info.py` and `video_snapshot.py` read this value from the environment |
| `TOS_SECURITY_TOKEN` | No | STS session token when using temporary credentials |
| `TOS_SAVEAS_BUCKET` | No | Default target bucket for saving snapshots |
| `TOS_SAVEAS_OBJECT_PREFIX` | No | Default key prefix for saving snapshots |
## Quick start (common tasks)
```bash
# Read video metadata (resolution, duration, codec)
TOS_OBJECT_KEY=demo.mp4 python3 {baseDir}/scripts/video_info.py
# Capture a single frame at 5 seconds
TOS_OBJECT_KEY=demo.mp4 python3 {baseDir}/scripts/video_snapshot.py --time 5000 --output frame_5s.jpg
# Capture multiple frames and save locally
python3 {baseDir}/scripts/video_snapshots.py --key demo.mp4 --timestamps 1000 3000 5000
# Capture multiple frames by interval
python3 {baseDir}/scripts/video_snapshots.py --key demo.mp4 \
--interval-ms 5000 --duration-ms 60000 --output-dir snapshots
# Capture multiple frames and save to TOS
python3 {baseDir}/scripts/video_snapshots.py --key demo.mp4 --timestamps 1000 3000 5000 --save-to-tos
# Concatenate multiple video clips
python3 {baseDir}/scripts/video_concat.py --key clip1.mp4 --fragments "clip2.mp4,clip3.mp4" --output-key merged.mp4 --wait
```
## Available scripts
| Script | Purpose |
|--------|---------|
| `scripts/video_info.py` | Read video metadata (format, duration, resolution, codec, frame rate). |
| `scripts/video_snapshot.py` | Capture a single frame at a given timestamp. Supports local save or TOS-to-TOS persistence. |
| `scripts/video_snapshots.py` | Batch-capture multiple frames across a timeline. Supports explicit timestamps, interval mode, concurrent execution, and TOS-to-TOS persistence. |
| `scripts/video_concat.py` | Concatenate multiple video clips into one. Uses TOS gateway `media_jobs` async API with `job_type=Concat`. Supports `--wait` to poll until completion. |
`video_snapshots.py` and `video_concat.py` support `--key`-style object selection. `video_info.py` and `video_snapshot.py` currently read `TOS_OBJECT_KEY` from the environment. `video_snapshots.py` also supports `--bucket` override for multi-bucket testing. Run any script with `-h` for full usage.
## Out of scope
- Full video transcoding pipelines (this skill covers metadata + frame extraction + concatenation only).
- Image-only or document-only workflows (use `byted-tos-image-process` or `byted-tos-doc-process`).
- Non-TOS storage providers.
## Rules
- **Authentication**: Authentication is provided by the TOS identity declared in the `metadata` block above. Object selection can be overridden per script with `--key`.
- **Timestamp unit**: All timestamp parameters are in **milliseconds** (e.g., `5000` = 5 seconds).
- **Validate timestamps**: Ensure snapshot timestamps do not exceed the video duration. Use `video_info.py` to read duration first when needed.
- **Save to TOS**: Use `--save-to-tos` with `TOS_SAVEAS_BUCKET` / `TOS_SAVEAS_OBJECT_PREFIX` to persist snapshots back to TOS without local download.
- **Parameter source of truth**: Official Volcengine TOS documentation is authoritative for the full `video/snapshot` parameter matrix. When uncertain, check [REFERENCE.md](REFERENCE.md).
- **Language**: Reply in the user's preferred language.
## Further reading
- Setup and environment: [README.md](README.md)
- Parameter reference: [REFERENCE.md](REFERENCE.md)
- End-to-end workflows: [WORKFLOWS.md](WORKFLOWS.md)
Related in Image & Video
watch
IncludedWatch a video (URL or local path). Downloads with yt-dlp, extracts auto-scaled frames with ffmpeg, pulls the transcript from captions (or Whisper API fallback), and hands the result to Claude so it can answer questions about what's in the video.
physical-ai-defect-image-generation
IncludedUse when the user wants to orchestrate defect image generation, run associated setup, or handle outputs on OSMO. The Day 0 path handles cold-start with USD-to-ROI, image-edit augmentation, and AnomalyGen to create initial PCBA datasets. The Day 1 path performs inference and labeling on real images. This skill helps with first-time asset setup, creation of finetuning checkpoints, and configuring deployment. Trigger keywords: defect image generation, dig workflow, dig pipeline, defect image detection workflow, aoi pipeline, aoi anomalygen, usd2roi anomalygen, day 0 pcba, day 1 pcba, day 1 real-photo alignment, day 1 manual roi, metal surface anomaly, glass defect, anomalygen finetune, setup_pcb, setup_metal, setup_glass, setup_pretrained, dig setup, dig datasets, dig pretrained checkpoint, dig image-edit endpoint.
accelint-react-best-practices
IncludedReact performance optimization and best practices. ALWAYS use this skill when working with any React code - writing components, hooks, JSX; refactoring; optimizing re-renders, memoization, state management; reviewing for performance; fixing hydration mismatches; debugging infinite re-renders, stale closures, input focus loss, animations restarting; preventing remounting; implementing transitions, lazy initialization, effect dependencies. Even simple React tasks benefit from these patterns. Covers React 19+ (useEffectEvent, Activity, ref props). Triggers - useEffect, useState, useMemo, useCallback, memo, inline components, nested components, components inside components, re-render, performance, hydration, SSR, Next.js, useDeferredValue, combined hooks.
elevenlabs-agents
IncludedBuild conversational AI voice agents with ElevenLabs Platform using React, JavaScript, React Native, or Swift SDKs. Configure agents, tools (client/server/MCP), RAG knowledge bases, multi-voice, and Scribe real-time STT. Use when: building voice chat interfaces, implementing AI phone agents with Twilio, configuring agent workflows or tools, adding RAG knowledge bases, testing with CLI "agents as code", or troubleshooting deprecated @11labs packages, Android audio cutoff, CSP violations, dynamic variables, or WebRTC config. Keywords: ElevenLabs Agents, ElevenLabs voice agents, AI voice agents, conversational AI, @elevenlabs/react, @elevenlabs/client, @elevenlabs/react-native, @elevenlabs/elevenlabs-js, @elevenlabs/agents-cli, elevenlabs SDK, voice AI, TTS, text-to-speech, ASR, speech recognition, turn-taking model, WebRTC voice, WebSocket voice, ElevenLabs conversation, agent system prompt, agent tools, agent knowledge base, RAG voice agents, multi-voice agents, pronunciation dictionary, voice speed control, elevenlabs scribe, @11labs deprecated, Android audio cutoff, CSP violation elevenlabs, dynamic variables elevenlabs, case-sensitive tool names, webhook authentication
humanizer
IncludedHumanize AI-generated text by detecting and removing patterns typical of LLM output. Rewrites text to sound natural, specific, and human. Uses 28 pattern detectors, 560+ AI vocabulary terms across 3 tiers, and statistical analysis (burstiness, type-token ratio, readability) for comprehensive detection. Use when asked to humanize text, de-AI writing, make content sound more natural/human, review writing for AI patterns, score text for AI detection, or improve AI-generated drafts. Covers content, language, style, communication, and filler categories.
generating-mermaid-diagrams
IncludedSalesforce architecture diagrams using Mermaid with ASCII fallback. Use this skill when generating text-based diagrams for Salesforce architecture, OAuth flows, ERDs, integration sequences, or Agentforce structure. TRIGGER when: user says "diagram", "visualize", "ERD", or asks for sequence diagrams, flowcharts, class diagrams, or architecture visualizations in Mermaid. DO NOT TRIGGER when: user wants PNG/SVG image output (use generating-visual-diagrams), or asks about non-Salesforce systems.