원클릭으로 Manus에서 모든 스킬 실행

veo-showreel-production-kit

스타192

포크33

업데이트2026년 7월 14일 20:05

Turn a video timeline + voiceover into a reproducible, sliceable Veo 3.1 prompt package with a consistency anchor and AI storyboard sketches.

설치

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

Manus에서 실행

출처

BlackBeltTechnology

BlackBeltTechnology/pi-agent-dashboard

GitHub 저장소 열기 Creator 저장소 보기

다운로드

Manus에서 실행

When to Use

Use when a user has a script/timeline + voiceover and wants to generate a consistent, professional AI video (Veo) where each fragment can be rendered separately/in parallel yet stays visually coherent and is easy to revise. Especially for trade-show/booth showreels.

Procedure

Research the target context (e.g., trade-show showreel best practices) and Veo prompting + consistency features (reference images, first/last-frame, seed, enhance_prompt=false). Write a research/strategy doc.
Write a STYLE BIBLE = global consistency anchor: single location/world, color palette, verbatim STYLE LOCK and AUDIO LOCK sentences, recurring object/character descriptions, a global NEGATIVE prompt, brand/compliance rules, and fixed reproduction settings (model, 16:9, 4K/24fps, constant seed, enhance_prompt=false).
Split the timeline into <=8s render units (Veo max clip ~8s); long scenes become A/B sub-shots. Mark each boundary as hard-cut (parallel) or SEAMLESS (chain last-frame of A as first-frame of B).
Generate per-shot markdown via a data-driven Python script: each file has the 7-layer prompt (camera, subject, action, environment, lighting, STYLE LOCK, AUDIO LOCK), an assembled Full Veo prompt (~110-170 words), the negative prompt, continuity notes, and repro settings. Also emit one combined VIDEO_MASTER.md.
Generate AI storyboard sketches with nano-banana (Gemini image model): a master world-anchor frame + one sketch per cut. Use these as Veo reference/first-frame images. Run with limited concurrency (ThreadPoolExecutor max_workers=3) over a sketch_prompts.json.
Audio policy: instruct Veo 'no spoken dialogue, no voiceover, no song vocals' so it only makes ambient SFX; the official VO + music are added in post.
Write an index README explaining read order, per-slice render steps, assembly, and how to revise one beat (re-render only that slice with same seed+anchor).

Pitfalls

Veo clips are ~8s max — never author a single >8s render unit; split into A/B and chain frames.
Paraphrasing the style block per shot causes drift; the STYLE LOCK and AUDIO LOCK must be byte-identical across every prompt.
Without a shared reference image attached to every generation, parallel slices diverge — always attach the world-anchor.
Let Veo invent narration if you don't explicitly forbid speech in the audio line.
For neutral/defense content add 'no flags, no national insignia, no real weapons, no brand logos' to the negative prompt and reserve a clean empty center for the logo in the final shot.
nano-banana CLI: npx -y @the-focus-ai/nano-banana "" --output file.png ; needs GEMINI_API_KEY; describe 16:9 in the prompt (square by default).

Verification

All shot files exist (one per cut) + VIDEO_MASTER.md + style bible + research + voiceover + README.
Every Full Veo prompt ends with the identical STYLE LOCK + AUDIO LOCK and references the same seed.
Storyboard sketches generated for the world anchor and every cut; spot-check key frames (opener, any on-screen-text shot, recurring character, final logo-space shot) for look + compliance.
Build a contact sheet (ImageMagick montage) to eyeball cross-shot consistency.

이 저장소의 다른 Skills

같은 저장소

ship-change

BlackBeltTechnology/pi-agent-dashboard

Ship an OpenSpec change after openspec-apply completes. When only QA/manual tasks remain, marks them done (tested later), archives + syncs specs, commits, pushes, opens a PR against develop, watches CI, waits for CodeRabbit, auto-applies safe fixes and re-pushes, loops until CI green + no actionable review threads, then squash-merges with branch delete and removes the worktree. Use after implementation is done and the change is ready to land. Triggers: "ship this change", "ship it", "land the change", "merge and clean up", "post-apply ship".

2026-07-14192

ship-it

BlackBeltTechnology/pi-agent-dashboard

Worktree-side implementation orchestrator for an OpenSpec change. Idempotent: gates automated scenarios on filesystem reality (test file exists + passes in the docker harness), not the tasks.md checkbox. Owns the red-test fix loop (never re-invokes apply on a checked task), runs the docker harness on its auto-derived port with always-teardown, then drives ship-change INLINE with a manifest-aware defer. Boundary-reverse escape hatch writes SHIP_IT_BLOCKED.md. Runnable headless. Triggers: "ship it", "build and ship this change", "run ship-it", "implement + test + land in the worktree".

2026-07-14192

switch-extension-source

BlackBeltTechnology/pi-agent-dashboard

Switch a monorepo extension/skill package between its published npm source and its local working-tree source, guaranteeing exactly one source per package. Use when "my edits don't take effect", "load local extension", "test the published npm build", or reconciling dev vs distribution package sources.

2026-07-14192

nano-banana-imagegen

BlackBeltTechnology/pi-agent-dashboard

Generate and edit images using Google Gemini image models via the nano-banana CLI. Use when the user asks to create, generate, make, or edit images with AI. Supports text-to-image, image editing, style transfer, and multi-image composition. Trigger on requests like "create an image", "generate a picture", "make me a logo", "edit this photo", "add X to this image".

2026-07-14192

veo-generator

BlackBeltTechnology/pi-agent-dashboard

Render a scripted video project into mp4 clips with the Google Veo 3.1 API. Reads a project's shot package (the shots/*.md format produced by the veo-showreel-production-kit) and generates one clip per camera cut — reusing the Full Veo prompt, negative prompt, seed, aspect/resolution, world-anchor reference and first-frame storyboard sketch already written in each shot file. Backed by the `pi-veo` CLI (a TypeScript port — no Python). API key is configurable per project via a .env. Use when: "render the Veo video for <project>", "generate the videos from the shot scripts", "make the clips with Veo", "re-render shot 03B".

2026-07-14192

distill-session-knowledge

BlackBeltTechnology/pi-agent-dashboard

Offline-mine this project's pi session JSONL logs into reusable, verified knowledge. Walks sessions newer than a watermark, extracts five signal classes (faults, ask_user decisions, corrections, procedures, docs) anchored on objective outcomes, promotes only patterns recurring across >= N sessions, and routes distilled artifacts into existing sinks (skill_manage, memory, docs + ctx_index). Use when the user says "mine my sessions", "distill session knowledge", "what have I learned across sessions", "extract lessons from logs", or "turn my pi history into skills/memory".

2026-07-14192

name	veo-showreel-production-kit
description	Turn a video timeline + voiceover into a reproducible, sliceable Veo 3.1 prompt package with a consistency anchor and AI storyboard sketches.
version	1
created	2026-06-08
updated	2026-06-08

When to Use

Procedure

Research the target context (e.g., trade-show showreel best practices) and Veo prompting + consistency features (reference images, first/last-frame, seed, enhance_prompt=false). Write a research/strategy doc.
Write a STYLE BIBLE = global consistency anchor: single location/world, color palette, verbatim STYLE LOCK and AUDIO LOCK sentences, recurring object/character descriptions, a global NEGATIVE prompt, brand/compliance rules, and fixed reproduction settings (model, 16:9, 4K/24fps, constant seed, enhance_prompt=false).
Split the timeline into <=8s render units (Veo max clip ~8s); long scenes become A/B sub-shots. Mark each boundary as hard-cut (parallel) or SEAMLESS (chain last-frame of A as first-frame of B).
Generate per-shot markdown via a data-driven Python script: each file has the 7-layer prompt (camera, subject, action, environment, lighting, STYLE LOCK, AUDIO LOCK), an assembled Full Veo prompt (~110-170 words), the negative prompt, continuity notes, and repro settings. Also emit one combined VIDEO_MASTER.md.
Generate AI storyboard sketches with nano-banana (Gemini image model): a master world-anchor frame + one sketch per cut. Use these as Veo reference/first-frame images. Run with limited concurrency (ThreadPoolExecutor max_workers=3) over a sketch_prompts.json.
Audio policy: instruct Veo 'no spoken dialogue, no voiceover, no song vocals' so it only makes ambient SFX; the official VO + music are added in post.
Write an index README explaining read order, per-slice render steps, assembly, and how to revise one beat (re-render only that slice with same seed+anchor).

Pitfalls

Veo clips are ~8s max — never author a single >8s render unit; split into A/B and chain frames.
Paraphrasing the style block per shot causes drift; the STYLE LOCK and AUDIO LOCK must be byte-identical across every prompt.
Without a shared reference image attached to every generation, parallel slices diverge — always attach the world-anchor.
Let Veo invent narration if you don't explicitly forbid speech in the audio line.
For neutral/defense content add 'no flags, no national insignia, no real weapons, no brand logos' to the negative prompt and reserve a clean empty center for the logo in the final shot.
nano-banana CLI: npx -y @the-focus-ai/nano-banana "" --output file.png ; needs GEMINI_API_KEY; describe 16:9 in the prompt (square by default).

Verification

All shot files exist (one per cut) + VIDEO_MASTER.md + style bible + research + voiceover + README.
Every Full Veo prompt ends with the identical STYLE LOCK + AUDIO LOCK and references the same seed.
Storyboard sketches generated for the world anchor and every cut; spot-check key frames (opener, any on-screen-text shot, recurring character, final logo-space shot) for look + compliance.
Build a contact sheet (ImageMagick montage) to eyeball cross-shot consistency.