← 返回 Skills 市场
Tracked Video Analysis
作者
Иван Романенко
· GitHub ↗
· v1.0.0
· MIT-0
408
总下载
0
收藏
2
当前安装
1
版本数
在 OpenClaw 中安装
/install tracked-video-analysis
功能描述
Analyze local or linked video files and convert them into structured summaries of features, functions, workflows, or topics. Use when a user wants a walkthro...
安全使用建议
This skill appears to do what it says: chunk a local video, run ASR, and produce a structured summary. Before running, ensure you: (1) install required npm packages (/@xenova/transformers, ffmpeg-static, ffprobe-static, wavefile) in an isolated environment; (2) expect the transformers library to download ASR model weights (network access and disk space required); (3) provide the video locally or a direct download link as instructed; (4) review and run the scripts in a workspace directory (they write tmp/status/log/transcript/final_analysis files) and avoid feeding sensitive videos unless you trust the environment and any remote model provider; and (5) if you need stricter network control, confirm where the transformer model is sourced from (Xenova/Hugging Face) before allowing runtime downloads.
功能分析
Type: OpenClaw Skill
Name: tracked-video-analysis
Version: 1.0.0
The skill bundle provides a legitimate framework for transcribing and summarizing video files using a two-stage process (extraction and structuring). It utilizes standard libraries such as @xenova/transformers for ASR and ffmpeg-static for media processing, with all operations confined to a local working directory (tmp/video_analysis/). The code in scripts/transcribe_tracked_light.mjs and scripts/final_structurer.py is transparent, lacks obfuscation, and contains no indicators of data exfiltration, unauthorized network access, or malicious intent.
能力评估
Purpose & Capability
Name/description match the provided artifacts: a JS extraction script and a Python structurer that read a local video (video.mp4 or workspace path), produce transcript.jsonl and final_analysis.md, and implement chunked ASR and grouping. There are no unrelated credential requests or unrelated binaries declared.
Instruction Scope
SKILL.md and references/pipeline.md limit actions to local acquisition, chunked ASR, and structured summarization and instruct the agent to read/write files under tmp/video_analysis or the working dir. The included scripts follow this pattern and write status/progress/transcript files. Note: the JS extraction script uses @xenova/transformers pipeline('automatic-speech-recognition', 'Xenova/whisper-tiny'), which will likely download model weights or perform network activity via that library at runtime; this is expected for ASR but should be considered when running offline or in restricted environments.
Install Mechanism
There is no install spec (instruction-only), but the code requires npm packages (@xenova/transformers, ffmpeg-static, ffprobe-static, wavefile) and Python for the structurer. This is not malicious but means the maintainer expects the runtime environment to install dependencies; model artifacts may be fetched by the transformers library at runtime. No arbitrary URL downloads or obscure extract/install steps are present in the skill itself.
Credentials
The skill requires no environment variables or credentials and the scripts do not reference secrets or external config paths. They only access local files (video.mp4, chunk_*.wav, status/log/transcript files).
Persistence & Privilege
Skill is not always-enabled and does not request elevated privileges. It writes progress and result files in the working directory and does not modify other skills or global agent configuration.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install tracked-video-analysis - 安装完成后，直接呼叫该 Skill 的名称或使用
/tracked-video-analysis触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial release: reliable two-stage local video analysis with tracked extraction, tracked structuring, status files, and structured summaries from noisy video sources.
元数据
常见问题
Tracked Video Analysis 是什么？
Analyze local or linked video files and convert them into structured summaries of features, functions, workflows, or topics. Use when a user wants a walkthro... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 408 次。
如何安装 Tracked Video Analysis？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install tracked-video-analysis」即可一键安装，无需额外配置。
Tracked Video Analysis 是免费的吗？
是的，Tracked Video Analysis 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Tracked Video Analysis 支持哪些平台？
Tracked Video Analysis 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Tracked Video Analysis？
由 Иван Романенко（@mrgoodgreen）开发并维护，当前版本 v1.0.0。
推荐 Skills