본문으로 건너뛰기
cccskills
무료GitHub에 공개

rn-replica-qc

Analyze, recreate, align, and verify reference-video motion. Use when the user asks to recreate, copy, match, align, pixel-align, or compare a video; when a HyperFrames or Remotion remake is judged not aligned; when fixed reference playback and reusable motion components must be separated; or when delivery needs full-frame runtime, render, decode, boundary, timing, and archive evidence instead of sampled screenshots alone.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

설치 방법 보기

포함된 파일(13)

  • SKILL.md6.9 KB
  • agents/openai.yaml364 B
  • assets/showcases/presenton-replica-pixel-aligned-bitexact.mp45.3 MB
  • references/alignment-workflow.md4.4 KB
  • references/presenton-lessons.md2.8 KB
  • references/replica-levels.md2.5 KB
  • scripts/build_frame_sprite_rows.py3.1 KB
  • scripts/compare_videos.py9.4 KB
  • scripts/extract_frame_range.py3.8 KB
  • scripts/extract_halfsec_frames.py5.9 KB
  • scripts/verify_frame_sequence.py9.5 KB
  • scripts/verify_frame_sprite_rows.py3.2 KB
  • scripts/verify_timeline_playback.cjs4.2 KB

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

Reference Video Replica QC

Role

Turn reference-video recreation into a red/green evidence loop and, when useful, capture the aligned motion as reusable HyperFrames or Remotion components.

Read replica-levels.md before naming or promising fidelity. A source-stream copy, an exact lossless render frame, an encoded frame-aligned MP4, a visual rebuild, and a style reference are different claims.

Canonical Showcase

  • Human display name: Presenton 复刻 Bitexact 成片
  • File: assets/showcases/presenton-replica-pixel-aligned-bitexact.mp4
  • This showcase demonstrates source-stream bit exactness. Do not use it as proof that a hand-authored HTML composition is bit exact.

Operating Modes

  • Analysis-only: inspect and report; do not edit implementation code.
  • Replica-loop: repair the candidate until the declared evidence gate passes.
  • Component-capture: register a proven motion pattern with inputs and limits.

Workflow

1. Declare The Contract

Record the reference, exact segment, renderer, fidelity label, deliverables, component-library destination, and evidence required for approval. Use Goal mode when available; otherwise keep alignment-report.md and patch-log.md beside the project.

Complete when: the segment and every acceptance claim can be tested without relying on chat memory.

2. Lock The Timebase

Probe FPS, duration, start time, pixel format, frame count, and audio presence. For exact work, map timestamps to source frame numbers and extract by frame index. Inclusive segment count is (end_frame - start_frame) + 1. When joining adjacent inclusive segments, keep the earlier boundary frame and skip the first frame of the next segment.

Use extract_halfsec_frames.py for discovery and extract_frame_range.py for full-frame acceptance sequences.

Complete when: reference frame 1, final frame, count, and PTS mapping are explicit and reproducible.

3. Choose The Architecture

  • Exact replay: fixed reference content rendered from verified lossless frames or sprite rows. It is not a freely generative component.
  • Parametric motion: DOM/SVG/GSAP/React/Three.js with replaceable content.
  • Hybrid: ship exact replay for fidelity and keep a separate parametric implementation for later abstraction.

Never present an exact-replay texture layer as a parameterized AI component.

4. Analyze And Repair

Read alignment-workflow.md. Start with a 0.5s contact sheet, add 0.1s windows around fast motion, then repair the earliest or largest mismatch one visible cause at a time. Sampling locates problems; it does not approve pixel or frame-aligned delivery.

For the Presenton failure modes and the dynamic-image decode race, read presenton-lessons.md.

Complete when: the candidate reaches the declared visual state and no known earlier mismatch remains hidden behind a later fix.

5. Pass Three Evidence Gates

  1. Asset gate: reference extraction count is exact; generated sprite cells or lossless intermediates round-trip to their source frames with zero error.
  2. Runtime gate: the real registered timeline advances sequentially through every required frame. Do not call a test-only setter. Check arbitrary seeks, every dynamic-asset boundary, and previously aligned neighboring segments.
  3. Delivery gate: run the renderer, decode the delivered MP4, compare every frame, check temporal candidates, inspect worst and boundary frames, probe media metadata, and verify archive copies by SHA-256.

For HyperFrames, npx hyperframes check . must pass before render. A successful render log is not delivery evidence. For Remotion, use the project's equivalent runtime and decoded-frame checks.

Complete when: all three gates are green and the report states which layer each metric proves.

6. Extend And Join

When continuing a replica in segments, rerun the previous segment, render the new segment independently, skip duplicate inclusive boundary frames, and check the frame before, at, and after every join. Then probe the combined frame count and duration.

Complete when: both old and new ranges pass and no join repeats, drops, or rewinds a frame.

7. Capture Components

Register exact replay and parametric motion separately. Each entry includes:

  • type: exact-replay or parametric-motion
  • source and timestamp range
  • intended content pattern
  • inputs and timing contract
  • implementation stack and assets
  • evidence files and acceptance level
  • known limits

8. Decide

Approve only the fidelity label proved by current evidence. If a gate is red, say not aligned, name the first failing frame or timestamp, and continue the replica loop.

Script Usage

# Discovery samples and contact sheets
python3 scripts/extract_halfsec_frames.py reference.mp4 --out analysis/sample --contact

# Exact inclusive frame range
python3 scripts/extract_frame_range.py reference.mp4 --start-frame 720 --end-frame 1020 --out analysis/reference-fullfps

# Lossless sprite rows and round-trip proof
python3 scripts/build_frame_sprite_rows.py analysis/reference-fullfps assets/textures --prefix replica-row --frames-per-row 7
python3 scripts/verify_frame_sprite_rows.py analysis/reference-fullfps assets/textures --prefix replica-row --frames-per-row 7

# Real timeline capture, followed by full-frame comparison
node scripts/verify_timeline_playback.cjs --html ./index.html --timeline-id composition_id --start 24 --frames 301 --fps 30 --out analysis/runtime
python3 scripts/verify_frame_sequence.py analysis/reference-fullfps analysis/runtime --output analysis/runtime-metrics.json --boundary-every 7

# Delivered video comparison and decoded-frame gate
python3 scripts/compare_videos.py reference.mp4 candidate.mp4 --out analysis/delivery --reference-start-frame 720 --candidate-start-frame 0 --frame-count 301 --boundary-every 7

Use a Python runtime with Pillow for frame and sprite verification. Timeline capture requires Playwright in the Node runtime.

Required Outputs

  • alignment-report.md and patch-log.md
  • source and candidate contact sheets
  • full-frame runtime and delivery metrics JSON
  • worst-frame and boundary-frame evidence
  • side-by-side comparison video or sheet
  • media probe and checksum evidence
  • component catalog entry when requested
  • final pass/fail summary

For content-creation projects, copy final videos into 成片/ and the compact QC evidence into the sibling 质检/ directory.

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

댓글

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

보내면 바로 공개됩니다. 이메일 주소나 전화번호 같은 개인 정보는 쓰지 마세요. URL은 2개까지입니다.

0 / 1,000자

이 양식은 reCAPTCHA로 보호되며 Google 개인정보처리방침 및 서비스 약관이 적용됩니다.

같은 저장소의 스킬

口播视频转录和口误识别。生成审查稿和删除任务清单。触发词:剪口播、处理视频、识别口误

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Pluviobyte/rnskill1.7천2026년 9월 21일 업데이트

将法天象地、同源巨大法相或角色力量显现的视觉参考制作成连续关键帧,再使用用户选择或可用的图生视频工具分段生成、检查和合成动作视频,不绑定成片平台。适用于本体与法相同框、领域展开、凝实与同步出招;不负责整部小说改编或普通图片轮播。

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Pluviobyte/rnskill1.7천2026년 9월 21일 업데이트

口播基础素材包生成。转录口播、识别口误、生成审核页;用户确认后剪出新视频,Agent 再基于剪后视频重新转写、AI 校对字幕,输出后续口播成片可用的 source_cut.mp4 和 subtitles.srt。触发词:剪口播、处理口播素材、准备口播素材、识别口误、基础素材包

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Pluviobyte/rnskill1.7천2026년 9월 21일 업데이트

口播视频成片 Skill。把文章/口播稿/SRT、剪后视频和 HTML/图片素材串成分镜稿、时间线预览和最终 MP4;成片比例和动画风格从用户配置读取,动画默认使用小黑风格。触发词:口播成片、做分镜稿、时间线预览、合成口播视频、导出竖屏MP4

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Pluviobyte/rnskill1.7천2026년 9월 21일 업데이트

自进化 skills。记录用户反馈,更新方法论和规则。触发词:更新规则、记录反馈、改进skill

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Pluviobyte/rnskill1.7천2026년 9월 21일 업데이트

dbs

무료

dontbesilent 商业工具箱主入口。双模式:任务前路由(你的问题该用哪个 skill)+ 任务后导航(刚做完诊断,下一步该干什么)。 触发方式:/dbs、/商业、「帮我看看」、「下一步怎么走」 Main entry point for dontbesilent business toolkit. Dual mode: pre-task routing + post-task navigation. Trigger: /dbs, "help me with my business", "what's next"

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

Pluviobyte/rnskill1.7천2026년 9월 21일 업데이트

Pluviobyte의 스킬 모두 보기

이 스킬의 문제 신고