Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/whitetowerai/cut-as-code/video-cutnpx skills add WhiteTowerAI/cut-as-code --skill video-cutgit clone --depth 1 https://github.com/WhiteTowerAI/cut-as-codeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/whitetowerai/cut-as-code/video-cut)<a href="https://agentmods.dev/skills/whitetowerai/cut-as-code/video-cut"><img src="https://agentmods.dev/badge/skills/whitetowerai/cut-as-code/video-cut.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00160 | $0.03672 |
| Opus 5 | $0.00080 | $0.01836 |
| Sonnet 5 | $0.00032 | $0.00734 |
| Haiku 4.5 | $0.00016 | $0.00367 |
Grade A, and why
video-cut scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 264 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Video Cut
Turn one unedited long video into a compact, watchable first cut — decisive, content-aware, and self-verified. Human (you, the model) makes the editorial decision of WHAT to keep; the bundled scripts do the mechanical precision (boundary alignment, dead-air reclaim, memory-safe render, self-check).
When to use
- A raw/unedited video (YouTube URL or local file) needs to become a tighter cut.
- The ask is "halve it", "cut the boring parts", "make a first/rough cut".
- NOT for: word-by-word filler micro-surgery, multi-cam B-roll assembly, color/audio polish — those are a later P1/P2 polish pass. This skill is P0: validate the diagnosis + produce a solid cut.
Dependencies
/video-understand is a prerequisite. Run it first and use its validated media,
transcript, analysis, and timeline instead of recreating them in this skill.
Before starting, verify that it is installed. If it is not, warn the user that
this prerequisite is missing and stop before processing media.
Activate any environment that has these (do NOT assume a specific conda env name):
yt-dlp(download),ffmpeg/ffprobe(cut/render/probe) on PATH- Python with
faster-whisper(CPU works:device=cpu, compute_type=int8) Check first:yt-dlp --version,ffmpeg -version,python -c "import faster_whisper".
Project protocol workflow
Use these durable project files:
work/cut/edit-plan.json # hand-authored keep/drop decisions only
work/timeline.json # generated precision ranges + source/program mapping
review/01-cut/cut-summary.md
review/01-cut/timeline-map.png
review/01-cut/boundary-review.mp4
review/01-cut/full-proxy.mp4 # optional whole-program pacing review
Each canonical decision has a stable id, action (keep or drop), start_s,
end_s, reason, and semantic/transcript evidence_refs. Keep decisions chronological;
V1 does not support reordered or duplicated source clips.
Reuse the shared evidence layer rather than transcribing or diagnosing again:
What ships with it
14 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- examples/edit_coarse.json 6.2 KB
- examples/edit_final.json 50 KB
- reference/edit_coarse.schema.md 5.0 KB
- reference/pitfalls.md 5.3 KB
- scripts/analyze.py 4.7 KB runs code
- scripts/assign_speed.py 10 KB runs code
- scripts/build_edit.py 5.3 KB runs code
- scripts/check_project_protocol.py 2.2 KB runs code
- scripts/cut_render.py 4.1 KB runs code
- scripts/inspect_bounds.py 2.9 KB runs code
- scripts/selfcheck_frames.py 2.5 KB runs code
- scripts/transcribe.py 584 B runs code
- scripts/verify_cut.py 3.5 KB runs code
- tests/test_inspect_bounds.py 5.6 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 264 lines · 160 tokens per session scan A f7aadfb21d7e
video-cut is a skill published in the GitHub repository WhiteTowerAI/cut-as-code (11 stars, last pushed 12d ago), licensed MIT. It adds 160 tokens to every session and 3,672 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
smart-video-editor
把多段原始素材剪成一条成片——先用视觉模型看懂每段画面拍了什么、哪几秒可用，再决定取舍、顺序、时长、调色和 BGM，最后用 ffmpeg 渲染。适用于"我给你几段视频，帮我剪一条小红书/朋友圈/vlog"这类需求。当用户提供视频文件并希望剪辑、拼接、配乐、调色、转成竖屏时使用。不做 AI 生成视频。.
openstoryline-use
Use this skill when OpenStoryline is already installed and the user wants to start the local MCP/Web services, create or continue a session, send editing instructions, perform multi-turn re-editing, and verify rendered video outputs, as well as Chinese requests like “启动 OpenStoryline”, “把 OpenStoryline 跑起来”, “用…
create_profile_style_skill
【META SKILL】分析当前剪辑逻辑与风格，总结并生成一个新的可复用 Skill 文件，存入剪辑技能库。Analyze the current editing logic and style, summarize and generate a new reusable Skill file, and store it in the editing skill library.
subtitle_imitation_skill
【CAPABILITY SKILL】基于用户提供的参考文案样本，对视频素材内容进行深度文风仿写，生成风格化脚本。Based on user-provided reference text samples, the video material is deeply rewritten in terms of writing style to generate a stylized script.
premiere-pro-mcp
Install, verify, troubleshoot, and operate the Adobe Premiere Pro MCP server. Use when a user wants an agent to set up Premiere MCP, connect Claude Code/Codex/Claude Desktop, control Premiere, import media, build sequences, edit timelines, apply effects, or diagnose bridge issues.
ai_transition_editing_skill
【WORKFLOW SKILL】使用 AI 生成的转场将用户素材串联起来。.