Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/S3YED/appie-kitnpx agentmods add skills/s3yed/appie-kit/video-editing-proWrote 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/s3yed/appie-kit/video-editing-pro)<a href="https://agentmods.dev/skills/s3yed/appie-kit/video-editing-pro"><img src="https://agentmods.dev/badge/skills/s3yed/appie-kit/video-editing-pro.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 4 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium MCP Rug Pull · line 630 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 639 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 668 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
- medium MCP Rug Pull · line 669 npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.Fix: Pin the version: npx @scope/[email protected]
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 5.1 | $0.00029 | $0.10476 |
| Opus 5 | $0.00015 | $0.05238 |
| Sonnet 5 | $0.00006 | $0.02095 |
| Haiku 4.5 | $0.00003 | $0.01048 |
Grade A, and why
video-editing-pro scanned grade A with 1 finding 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 4d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -L "https://cdn.mixkit.co/sfx/download/mixkit-fast-small-sweep-transition-166.wav" -o sfx/whoosh-1.wav How it starts
The opening of the file, as written. The whole thing — 690 lines — stays where its author put it; the contents beside it link to each section on GitHub.
video-editing-pro Skill
Purpose
Decision-level framework for cutting talking-head short-form video (Reels/Shorts/TikTok). Not tool instructions — cut decisions: which sentences survive, where to end them, how to pick the hook, how to pace energy, and how to know when you got it wrong. Codified from auto-editor source defaults, production-audio practitioner interviews, platform retention data, and analysis of the 2026-05-03 content factory clips.
Source
Every entry consumed directly, not cited generically:
| Source | URL | What it gave |
|---|---|---|
| auto-editor reference | https://auto-editor.com/ref/edit | Exact defaults: threshold=0.04 (normalized amplitude), margin=0.2s padding, smooth=0.2s,0.1s |
| auto-editor options | https://auto-editor.com/options | --margin 0.2s default, --smooth 0.2s,0.1s (mincut 0.2s, minclip 0.1s) |
| auto-editor blog: threshold removal | https://auto-editor.com/blog/silent-threshold | Why 0.04 is the amplitude default, not dB |
| Production Expert: René Coronado interview edit | https://www.production-expert.com/production-expert-1/2020/7/29/editing-audio-interviews-what-choices-do-we-make-and-why | "My goal is NOT to get from 14s to 6s" — clarity-of-thought frame vs duration-reduction frame |
| podmuse.com: pro podcast edit workflow | https://www.podmuse.com/post/edit-audio-for-podcast | Selective filler removal rule: cut fillers that block clarity or stack awkwardly, leave ones that "sound human and disappear in context" |
| Recut silence removal docs | https://getrecut.com/remove-silence-from-video-automatically | Padding 0.4–0.5s sounds natural; 0s creates rapid-fire effect |
| Aibrify: YouTube Shorts retention curve playbook 2026 | https://aibrify.com/blog/youtube-shorts-retention-curve-playbook | Pacing target 1.5–2s per visual change; below 1.2s = noise, above 2.5s = dip; 55–70% retention for talking-head |
| TikTok editing guide | https://edicionvideopro.com/en/editing-for-platforms-video-marketing/tiktok-video-editing-guide/ | "Every pause between sentences, every transitional phrase should be cut"; drop-off graph as diagnostic |
| Postigniter: hook formulas | https://postigniter.com/blog/high-performance-hooks-writing-scroll-stopping-intros-for-short-form-video | Optimal hook 1.5–3s; 0.1–0.3s cuts for dynamism in hook window |
| go-viral.app: first 3 seconds | https://www.go-viral.app/blog/hook-first-3-seconds/ | <50% retention at 3s = hook failing; >70% = strong; target >70% |
| autoshorts.app: hooks 2026 | https://www.autoshorts.app/blog/instagram-reels-hooks-2026 | 3+ cuts in first 3s boost watch time 58%; each cut creates "change signal" |
| VlogLikePro: energy management | http://vloglikepro.com/editing-styles-that-drive-watch-time-beyond-basics | Energy curve; cut every 4s→3s→2s building to reveal; 10–15s micro-hook cadence |
| We Design Motion: attention without faster cuts | https://wedesignmotion.com/blog/design/video-content-that-earns-attention-without-faster-cuts/ | 5-layer attention model: promise→clarity→progression→contrast→craft |
| socialync.io: hook-body-payoff structure | https://www.socialync.io/blog/short-form-video-structure-guide-2026 | "8 seconds without visual change = viewer exit"; tension > rush to resolution |
| Ascynd: Hormozi captions breakdown | https://ascynd.io/en/blog/hormozi-captions | One keyword per phrase; noun or verb carrying meaning; yellow #FFD93D; 200–500ms per word timing |
| reelwords.ai: caption retention | https://reelwords.ai/blog/how-to-add-captions-to-short-form-video | Highlight: promise words, contrast words, specificity (numbers/outcomes), stakes words; never highlight "the", "and", "like" |
| Film Editing Pro: reviewing your own cuts | https://filmeditingpro.com/3-editing-tips-for-reviewing-your-own-cuts | Audio-off test; black-and-white test; Hitchcock rule; first-reaction markers |
| joelv.ca: post-edit fresh-eyes technique | https://joelv.ca/blog/best-way-to-edit-better | Watch with someone beside you = best fresh-eyes; Joe Walker (Dune) black-and-white review trick |
| autocut.com: finding viral moments | https://autocut.com/en/blogs/viral-moment/ | Three criteria for clip viability: hook (curiosity), clarity (standalone), emotion (tension/surprise/reaction) |
What ships with it
5 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.
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.
- 4d ago First seen · 690 lines · 29 tokens per session scan A 9751d6f271b2
video-editing-pro is a skill published in the GitHub repository S3YED/appie-kit (7 stars, last pushed 12d ago), licensed MIT. It adds 29 tokens to every session and 10,476 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
hive.image-generation
Required before calling imagegenerate. Create and edit images from a prompt — generate an image, make a picture / logo / illustration / icon / banner / poster / thumbnail / hero image / mockup / product shot / social graphic, or edit / restyle / combine existing images from reference images. Uses OpenAI gpt-image-2…
shogun-screenshot
A screenshot tool for getting images from a computer or web page and then cropping, resizing, or masking sensitive information. Playwright is a browser-automation tool used here to capture web pages.
agent-tamagotchi
Terminal pet that lives in your statusline. 12 species, 5 stats (DEBUGGING, PATIENCE, CHAOS, WISDOM, SPEED). Reacts to your workflow - happy when tests pass, sad when builds fail, excited during swarm mode. Deterministic species from user ID.
product-showcase
Generate a comprehensive marketing website for a web app — multi-page with real screenshots, animated GIF walkthroughs, feature deep-dives, and workflow demonstrations. Browses the running app, captures screens and sequences, and produces a deployable site that actually teaches people what the product does. Especially…
audio-production
Professional audio production for music, podcasts, and sound design. Use when working with audio recording, mixing, mastering, or sound design for any medium.
brand-identity
Brand strategy and identity design for businesses and products. Use when creating brand guidelines, developing visual identity systems, or defining brand positioning.