闲社服务运行正常AI智能体自动化平台
l

long-video-short-clip-selector

Use when a creator or content team needs to rank time-coded transcript moments, block risky clips, and render reviewable rough cuts from an owned or licensed long video.

作者:暂未提供 | 来源记录:ClawHub
包含在套餐中
登记来源
ClawHub
版本
V 1.0.0
检测标记
后台标记通过
随套餐获取
请查看所属套餐
0
收藏
来源与检测标记为平台登记信息,并不代表已展示可复核的检测报告。使用前请核对版本、依赖和所需权限,建议先在隔离环境中运行。
概述
安装方式
版本历史

long-video-short-clip-selector

Long Video Short Clip Selector

Procedure

  1. 1. Confirm source rights and speaker consent, then map a timestamped transcript to references/transcript-contract.md.
  2. Run scripts/select_short_clips.py with an approved scoring policy and previous state.
  3. Review selected moments and risk flags, then use scripts/render_clips.py to create local rough cuts for editorial approval.

Guardrails

  • - Process only video owned by the user or covered by a license that permits clipping.
  • Never publish, upload, schedule, or claim performance automatically.
  • Block clips containing personal contact data, unsupported numeric claims, unverified rights, or missing consent.
  • Preserve source timecodes, score breakdown, risk reasons, and input hashes.
  • Treat a high score as an editing priority, not a promise of views or virality.

Outputs

The workflow creates ranked candidates, a non-overlapping selection, review queue, rights report, platform copy drafts, FFmpeg manifest, audit log, and next-run state. The optional renderer produces local MP4 rough cuts only.

标签

video shorts transcript ffmpeg content-repurposing n8n

下载

v1.0.0 最新 2026-8-20 00:44
初始版本(来自套餐拆分)
返回顶部