本文へ移動
cccskills
無料GitHub で公開

Video Effects

Apply a short artistic effect or VFX transformation to an approved media parent with explicit intent, provenance, and safety labeling.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md3.0 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Video Effects

Mission

Transform an approved image or short video with an artistic effect—such as freeze-time, stylization, orbit, distortion, particles, or a VFX reveal—while keeping the parent asset and synthetic nature visible. Effects are for creative entertainment and post-production drafts, not factual restoration or evidence.

Safety and scope gate

Confirm rights for the source, people, brands, music, and location. Decline non-consensual likeness manipulation, deceptive disaster/news footage, political misinformation, exploitative content, and harmful edits involving minors. Do not use an effect to remove a watermark or conceal ownership.

Collect source URL/path, effect intent, target duration, ratio, resolution, quality, audio, preservation rules, and desired intensity. The current public MuAPI effects guidance emphasizes short clips and prompt/preset-driven transformations; verify the live schema before choosing an effect name or input type.

Execute

  1. Freeze the parent and describe the effect as a visible temporal change: start state, trigger, transformation, end state, camera, lighting, and sound. Avoid vague “make it cool” prompts.
  2. Upload a local source. If the selected effect endpoint requires an image reference and the user has not supplied one, stop and request an approved still; this pack does not assume a cross-transport frame-extraction tool.
  3. Resolve a live effect/VFX schema using MODELS.md. The documented MuAPI REST family uses generate_wan_ai_effects with effect/name, prompt, reference, ratio, resolution, quality, and short duration fields when available. Use media.edit_video as the logical operation and do not assume a generic MCP alias exists.
  4. Submit one proof, retain prediction ID, poll the result, and compare parent and child throughout the effect. Check flicker, subject boundaries, unwanted identity changes, physics, audio, and output retention.
  5. Label the output as artistic/generated and run video QA. Return the parent if the effect obscures the subject or introduces unacceptable defects.

Output contract

Return parent ID, effect intent/name, model/endpoint, exact live fields, prediction ID, output URL, status, synthetic label, audio treatment, QA findings, and failed/unsupported branches. Save permitted receipts under .video/ using reports-and-receipts.md.

Effects are draft-only and never certify a real event, restoration, or brand representation.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Plan draft video-ad variations from supplied performance evidence, preserving the distinction between observed data and creative hypotheses.

日本語の概要は準備中です。原文の説明を表示しています。

SamurAIGPT/open-ai-video-agent72026年8月28日 更新

Create authorized presenter, avatar, and UGC-style video drafts with explicit voice, likeness, consent, and synthetic-media handling.

日本語の概要は準備中です。原文の説明を表示しています。

SamurAIGPT/open-ai-video-agent72026年8月28日 更新

Analyze a script or existing edit, identify visual gaps, and generate supplemental B-roll coverage with clear parent and factual-status labels.

日本語の概要は準備中です。原文の説明を表示しています。

SamurAIGPT/open-ai-video-agent72026年8月28日 更新

Build a consented continuity system for an authorized person, fictional character, product, or world across video shots.

日本語の概要は準備中です。原文の説明を表示しています。

SamurAIGPT/open-ai-video-agent72026年8月28日 更新

Translate creative intent into coherent cinematography, framing, camera, lens, lighting, motion, and sound direction for a video shot.

日本語の概要は準備中です。原文の説明を表示しています。

SamurAIGPT/open-ai-video-agent72026年8月28日 更新

Create a non-explicit romantic scene from authorized adult references with clear consent, identity boundaries, and synthetic-media labeling.

日本語の概要は準備中です。原文の説明を表示しています。

SamurAIGPT/open-ai-video-agent72026年8月28日 更新

SamurAIGPT のスキルをすべて見る

このスキルの問題を報告する