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/runesleo/claude-video-kitnpx agentmods add skills/runesleo/claude-video-kit/video-explainerWrote 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/runesleo/claude-video-kit/video-explainer)<a href="https://agentmods.dev/skills/runesleo/claude-video-kit/video-explainer"><img src="https://agentmods.dev/badge/skills/runesleo/claude-video-kit/video-explainer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/runesleo/claude-video-kit/video-explainer"><img src="https://agentmods.dev/badge/skills/runesleo/claude-video-kit/video-explainer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, 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 18 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 | $0.00072 | $0.01989 |
| Opus 5.5 | $0.00029 | $0.00796 |
| Sonnet 5.5 | $0.00014 | $0.00398 |
| Haiku 4.5 | $0.00007 | $0.00199 |
Grade A, and why
video-explainer 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 15d 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 — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Video Explainer
Version note: v0.3.0-rc.1 / 2026-09-27 / Make production strategy adaptive while keeping the deterministic Remotion recipe as a strict fallback. Final-artifact acceptance is strategy-independent and bound to the exact MP4 bytes. Keep upload, account access, credentials, private voice assets, hosted rendering, and publication out of scope.
Produce a local video and evidence packet. Never upload or publish from this skill.
Core contract
This Skill separates what must be true of the final artifact from how the artifact is produced.
Before forcing an existing recipe, run a capability challenge:
- Can the current model or available capability produce the final MP4 directly at the required quality?
- Which requirements are real product, safety, brand, or evidence invariants?
- Which existing steps only compensate for older model limitations?
- Can intermediate stages be skipped without weakening final acceptance?
- What is the simplest strategy likely to pass the final gates?
- If that strategy fails, fall back to the deterministic repository-owned recipe.
Choose one execution strategy:
direct_artifact— a current model/capability produces the final MP4 directly. Do not requiremotion-storyboard.json,script.json, TTS, alignment, or Remotion merely to satisfy a historical recipe.agent_orchestrated— an agent may choose and combine tools as implementation details. Intermediate artifacts are optional unless they are needed to satisfy the brief or evidence requirements.legacy_remotion— use the repository-owned deterministic path when direct generation cannot preserve facts, branding, layout control, narration identity, or other required qualities.
All three strategies converge on objective MP4 verification plus an independent final-artifact review. Production freedom never bypasses final acceptance.
Install
The deterministic fallback orchestrates a local clone of claude-video-kit; it does not bundle the renderer. From the repository root, install the runtime and register the Skill with:
What ships with it
7 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.
- 15d ago Changed · +3 lines 2ccb677aa05f
- 25d ago Changed · +51 lines · -9 tokens per session f2fbe3e40d52
- 1mo ago Changed · +26 lines · +7 tokens per session 88a520a8a213
- 1mo ago First seen · 69 lines · 74 tokens per session scan A 53d9f2ab7ebf
video-explainer is a skill published in the GitHub repository runesleo/claude-video-kit (120 stars, last pushed 16d ago), licensed MIT. It adds 72 tokens to every session and 1,989 once invoked, about $0.0003 per session on Opus 5.5. 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
douyin-cover
A design workflow for creating or revising cover images for Douyin, WeChat Channels, Xiaohongshu, and other short-video content. It focuses on readable titles, clear layouts, and a recognisable creator identity.
music
Generate music using ElevenLabs Music API. Use when creating instrumental tracks, songs with lyrics, background music, jingles, or any AI-generated music composition. Supports prompt-based generation, composition plans for granular control, and detailed output with metadata. For workspace video BGM, search the…
viral-video-benchmark
A review workflow for Douyin and Xiaohongshu posts, two Chinese social-media platforms. It checks public performance information, compares posts with similar creators, and analyzes selected high-performing posts.
audio-to-subtitles
Convert local audio/video files or public media URLs into subtitle files by uploading local files to Cloudflare R2 and calling Volcengine AI MediaKit ASR subtitles API. Also supports MiniMax text-to-speech generation from text or text files, with optional subtitle generation from the produced audio. Supports batch…
media-to-transcript
Convert audio/video URLs or local media into corrected Markdown transcripts through Volcengine recording-file ASR 2.0. Reuses the existing video-transcript downloader for Bilibili, Douyin, Xiaohongshu, YouTube, extracts audio when input is video, uploads local audio to R2, splits media longer than 3 minutes into…
talking-head-hyperframes
A production skill for preparing and checking fixed talking-head video templates for HyperFrames, including a picture-in-picture presenter area. It archives inputs, checks media and subtitles, and creates a handoff for later scene work.