菜单
Find interesting moments in a transcript or video. Use when the user invokes /find-moments.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
One-shot conversion of a Claude Code plugin to Codex format. Use when asked to convert, port, or add Codex support for a plugin in this repository.
One-shot conversion of a Codex plugin to Claude Code format. Use when asked to convert a Codex plugin to Claude Code, add CC support for a Codex plugin, or reverse-port a plugin from Codex.
Continuous maintenance workflow for plugins that target both Claude Code and Codex. Use when a plugin's commands, skills, or agents have changed and the Codex target needs updating.
Manually sync the editor theme to the current macOS appearance (light/dark). Use when the user switched macOS appearance and wants Claude Code and/or Codex themes refreshed now, or asks to fix/resync the theme. Light → gruvbox-light, dark → sunset-drive.
Install the plugin-cross-port pre-commit hook into any git repository. Use when the user invokes /install-hook.
Generates platform-specific social media descriptions for video clips. Creates copy for YouTube Shorts, Instagram Reels, and TikTok.
| name | clip-maker-find-moments |
| description | Find interesting moments in a transcript or video. Use when the user invokes /find-moments. |
| version | 0.1.0 |
Converted from Claude Code command
/find-moments. Review and adapt: hooks and MCP tool IDs may need manual mapping for Codex.
Analyze a transcript (or video) to find the most engaging moments for vertical clips.
Parse $ARGUMENTS for:
<path> — required, path to either:
transcript.json — pre-existing transcript.mp4, .mov, .avi, etc.) — will transcribe first--duration N — target moment duration in seconds (default: 60)Check the file extension:
.json → use as transcript directlybash plugins/clip-maker/scripts/install-deps.sh
bash plugins/clip-maker/scripts/transcribe.sh "<video_path>" "<output_dir>"
Launch the moment-finder agent:
Agent(subagent_type="clip-maker:moment-finder", model="opus", prompt="
Analyze the transcript and find the most engaging moments.
- Transcript path: <transcript_path>
- Target duration: <duration> seconds
- Output path: <output_dir>/moments.json
Read the transcript, find the best moments, and write moments.json.
")
Read moments.json and present moments to the user as a formatted list: