video-understandUse when a video project needs reusable media metadata, word-level transcription, objective speech analysis, or evidence-backed semantic understanding before optional editing skills run.
Install via ClawdBot CLI:
clawdbot install sifr42/video-understandGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Sends data to undocumented external endpoint (potential exfiltration)
upload → https://platform.moonshot.ai/docs/api/filesCalls external URL not in known-safe list
https://www.youtube.com/watch?v=VIDEO_IDAI Analysis
The skill's external API usage (Gemini, Moonshot/Kimi) is consistent with its stated purpose of video analysis, and the UNKNOWN_DATA_SINK signal appears to be a documented file upload endpoint for the legitimate Moonshot AI provider. The UNDOCUMENTED_EXTERNAL signal for YouTube URLs is an expected input source, not a hidden call. No evidence of credential harvesting, intent override, or obfuscation was found in the provided definition.
Audited Apr 16, 2026 · audit v1.0
Generated Mar 21, 2026
Instructors can upload lecture recordings to generate summaries, extract key concepts, and create timestamped study guides. Students can ask follow-up questions about specific segments without re-watching the entire video.
Marketing teams analyze competitor YouTube ads or campaign videos to identify messaging, visual elements, and engagement tactics. They can query for specific details like product features or emotional cues to inform strategy.
Video editors use the tool to quickly review raw footage, generate scene descriptions, and locate clips based on queries (e.g., 'find all shots with a blue car'). This speeds up logging and editing processes.
Organizations analyze training or surveillance videos to verify adherence to protocols, such as safety procedures in manufacturing. The tool can flag deviations by answering questions like 'Are workers wearing helmets?' with timestamps.
Support teams upload tutorial or troubleshooting videos to create searchable knowledge bases. Agents can ask follow-up questions to quickly find solutions for customer issues without manually scanning content.
Offer tiered plans based on usage limits (e.g., videos per month, analysis duration). Target small businesses and enterprises needing regular video insights, with premium features like API access and custom integrations.
Provide the core functionality as an API for developers to integrate into their applications. Charge per video analysis or minute processed, appealing to startups and tech companies building video-centric products.
Sell customized on-premise or cloud deployments with enhanced security, support, and training. Focus on industries like media, education, or compliance where data privacy and high-volume processing are critical.
💬 Integration Tip
Use the --json flag for programmatic output and cache uploads to avoid redundant API calls, reducing costs and latency.
Scored Apr 19, 2026
视频自动剪辑助手。基于 FFmpeg 自动提取精彩片段、生成字幕、裁剪时长、制作短视视频，支持多平台导出。当用户需要：自动剪辑视频、提取关键词片段、生成字幕并烧录、导出短视频、提取视频精华、生成视频摘要时使用此技能。
Perform video editing tasks with ffmpeg, including cutting, merging, converting formats, extracting audio, adding subtitles, resizing, cropping, adjusting sp...
When the user wants to optimize signup, registration, account creation, or trial activation flows. Also use when the user mentions "signup conversions," "reg...
Use when the user has a video + a target-language SRT and wants the video to actually speak that language — generates a time-aligned TTS voice dub. Routes by...
Download videos, extract transcripts, capture frames. Analyze YouTube, tutorials, DD videos with yt-dlp + Whisper + ffmpeg.
Extract frames or short clips from videos using ffmpeg.