@autoaiai/aicut 0.1.11 → 0.1.13

package/skills/book-video-production/SKILL.md

@@ -1,211 +0,0 @@
----
-name: book-video-production
-description: "Research, script, storyboard, illustrate, cover-design, voice, caption, score, build, inspect, and render fact-based Chinese book short videos for Douyin, Xiaohongshu, TikTok, or other vertical platforms. Use when Codex is asked to create or reproduce a book-review, book-introduction, knowledge, or reading-account video; turn a book, reference video, narration, transcript, images, or MP4 into a vertical social video; choose book-specific vintage scenery or cartoon illustrations; create a publishing cover; add restrained background music; or establish a reusable AICUT or AICUT production pipeline. Runs in AICUT: the film is film.html, a page per shot, drawn by `aicut look` and `aicut render`."
-license: MIT
-source: "yuwenbin121/book-video-production-skill"
----
-
-# Book Video Production
-
-## Running it in AICUT
-
-AICUT's film is `film.html`: it cuts pages, footage, pictures and sound together on tracks. A page is a web page that
-sets `window.film = { frame(t) }` and draws that moment — no framework, no build step, no composition object, no
-`registerRoot`. Five shots are five pages cut together, not one page that changes scene, so each shot stays trimmable
-and replaceable in Studio.
-
-```sh
-npx -y @autoaiai/aicut@latest          # the manual: the format and the commands, always current
-npx -y @autoaiai/aicut@latest open .   # make this folder a film and print Studio's address
-npx -y @autoaiai/aicut@latest look     # a contact sheet of the whole film, its sound, and a determinism check
-npx -y @autoaiai/aicut@latest look 4.5 # one frame, full size — look at the PNG, not at the code
-npx -y @autoaiai/aicut@latest render   # the MP4, beside film.html
-```
-
-Run the manual before the first film: it is authoritative for the format, and this skill is about the method above it.
-
-Where the words below name another program's part, it becomes this:
-
-| elsewhere | in AICUT |
-|---|---|
-| composition, timeline, project file | `film.html`; tracks and clips inside it |
-| a scene, a sequence, a slide | a page (`pages/<name>.html`) cut onto a track |
-| `useCurrentFrame()`, `interpolate()` | `frame(t)` — the picture depends only on `t`, never on state or `Date.now()` |
-| a still, a screengrab | `aicut look <t>` |
-| `render` / `build` / `export` | `aicut render` (the person can also export from Studio) |
-| a dev server, a preview window | `aicut open` and the Studio address it prints |
-
-Hard rules that hold whatever the method says: the picture depends only on `t`; every value in the person's prompt is
-binding; look at the frames `look` writes before saying a film is done.
-
-
-Produce a reviewable video package first, then render only after factual, editorial, timing, and visual gates pass. Treat Codex as the coordinator and the user as final editor.
-
-## Establish the brief
-
-Resolve from the request or infer conservative defaults:
-
-- platform, duration, aspect ratio, language, audience, goal, and cover requirements;
-- book edition, author, cover, and allowed source material;
-- reference video or desired visual and vocal qualities;
-- desired deliverable: research, narration, storyboard, preview, final MP4, or reusable template.
-
-Use defaults of 30 seconds, 9:16, Chinese, knowledge-oriented book introduction, a separate publishing cover, quiet supporting music, and 4–6 scenes when unspecified. For Douyin, use 1080×1920 at 30 fps unless the user supplies another specification. Record assumptions and continue unless a missing choice would materially change the result.
-
-Create a production manifest using `references/production-schema.md`. Update it as decisions and asset paths become known.
-
-## Select the production route
-
-- Prefer HyperFrames for editorial typography, explainers, data motion, HTML/CSS layouts, and rapid visual iteration. Read and use the installed `hyperframes:hyperframes` and `hyperframes:hyperframes-cli` skills before authoring or invoking the CLI.
-- Prefer Remotion for React-based reusable series, strong data-driven templating, or repeated batch rendering. Read and use `remotion:remotion-best-practices` before writing Remotion code.
-- If the user explicitly chooses a route, honor it.
-- Do not install tools or plugins unless the user requests installation. If a required capability is unavailable, report the exact missing dependency and continue with upstream deliverables where possible.
-
-## Analyze references
-
-When a reference video is supplied:
-
-1. Inspect duration, dimensions, frame rate, audio, and scene changes with available media tools.
-2. Sample frames across the timeline and transcribe the audio when possible.
-3. Extract reusable structural traits: hook timing, information density, scene rhythm, image treatment, cover language, caption behavior, transitions, music role, and vocal qualities.
-4. Separate general techniques from protected expression. Reproduce pacing and design principles, not distinctive wording, artwork, branding, or a creator’s exact voice.
-
-Create a reference-style report before building. Include timestamps, representative frames, palette, typography, image categories, caption zones, transition vocabulary, music mood and approximate speech-to-music balance. Treat the report as a design constraint, not permission to copy protected assets.
-
-When the user asks for a quiet, premium literary treatment based on full-screen scenery and restrained captions, read `references/premium-literary-style.md` and adapt it to the specific book.
-
-If no reference video exists, use a restrained editorial knowledge-video structure.
-
-## Research the book
-
-Browse when the user requests research or when current or precise source attribution matters. Prefer:
-
-1. publisher, author, official edition, or library records;
-2. the supplied book text, notes, or licensed excerpts;
-3. reputable interviews and established reviews;
-4. reader comments and highlights only as attributed audience signals.
-
-For WeRead or another private service, use its connector only if available and authorized. Otherwise ask for exported highlights or omit that evidence. Never imply access that did not occur.
-
-Maintain an evidence table with claim, source, confidence, and intended use. Clearly distinguish:
-
-- verified bibliographic facts;
-- paraphrased ideas from the book;
-- attributed reviewer or reader opinions;
-- the scriptwriter’s interpretation.
-
-Do not invent quotations, page numbers, popularity, awards, reviews, or reader highlights. Use short quotations only when necessary and permitted.
-
-## Write and approve narration
-
-Write narration before building visuals. Read `references/editorial-style.md` and follow its drafting and audit rules.
-
-Structure a typical 30-second script as:
-
-1. a concrete curiosity or tension in the first 1–3 seconds;
-2. immediate identification of the book;
-3. one or two specific ideas or observations;
-4. a grounded reason the idea matters;
-5. a natural closing thought or light call to action.
-
-Estimate duration from the intended delivery rate, then revise to fit. Preserve breathing room. Label any uncertain claim instead of smoothing it into certainty.
-
-Deliver the narration with an evidence table and an editorial audit. Do not begin expensive rendering before the script is approved when the user is actively reviewing the project.
-
-## Generate and finish voice
-
-Use the user’s preferred TTS provider when available. Chinese pronunciation accuracy takes priority over timbre. Keep the raw generation and processed master separate.
-
-When given a reference voice, analyze transferable acoustic properties such as pace, pauses, pitch range, loudness, EQ balance, compression, and room tone. Do not claim to clone a person’s identity or promise an exact match.
-
-Apply only measured processing needed for clarity and tone. Check clipping, noise, intelligibility, pronunciation, and final loudness. If the user supplies an MP4 containing narration, extract the audio losslessly where possible.
-
-Treat the final processed narration as the timing source of truth.
-
-## Transcribe and storyboard
-
-Transcribe the final narration rather than guessing subtitle timing. Use word- or phrase-level timestamps when the available transcriber supports them.
-
-Build 4–6 scenes around semantic beats. For every scene specify:
-
-- exact start and end;
-- matching narration phrase;
-- visual purpose and composition;
-- on-screen copy and highlighted keyword;
-- asset source;
-- animation and transition;
-- caption-safe region.
-
-Do not place a visual cut merely at a fixed interval; align it to meaning, emphasis, or pause. Keep the cover readable and avoid turning every spoken sentence into duplicate on-screen prose.
-
-## Design the cover and source visuals
-
-Read `references/visual-assets-and-cover.md`. Produce a separate 9:16 publishing cover in addition to the video opening frame.
-
-Derive the visual world from the individual book rather than applying one fixed aesthetic. Select among archival or vintage scenery, period objects, documentary textures, maps, symbolic still life, painterly illustrations, or cartoon images according to genre, setting, era, emotional register, and audience.
-
-Build an asset plan mapping each visual to a narration beat and source. Prefer user-owned, public-domain, properly licensed, or newly generated assets. Do not use a published book cover, film/TV still, illustrator’s recognizable style, or copyrighted character unless authorized. If generating images, keep characters, palette, era, and aspect ratio consistent across scenes.
-
-Make the cover readable at phone-thumbnail size. Include the book title as the primary text, one short curiosity line or emotional promise, and a clear focal image. Avoid misleading clickbait, dense synopsis text, and platform UI collision.
-
-## Add restrained background music
-
-Read `references/music-mix.md`. Choose or generate music only after the narration’s emotional arc is known.
-
-Use a light, slow supporting bed by default. Keep speech clearly dominant; use the music to establish atmosphere and bridge pauses, not to create a second focal point. Duck music under narration, fade cleanly at both ends, and verify the final mix for clipping, intelligibility, and sudden loudness changes.
-
-## Build the composition
-
-Create a deterministic project with local or properly licensed assets. Keep content data separate from the visual template so another book can reuse the composition.
-
-For HyperFrames:
-
-- author valid scene elements with correct `data-start`, `data-duration`, and `data-track-index`;
-- derive timing from the transcript and final audio duration;
-- lint, inspect, and preview with the HyperFrames CLI skill;
-- verify the exact installed CLI syntax rather than guessing flags.
-
-For Remotion:
-
-- store book metadata, scenes, captions, and assets in typed data;
-- derive duration and sequences from timestamps;
-- follow the installed Remotion skill for current APIs, animation, media, and rendering.
-
-Use licensed fonts that contain all required Chinese glyphs. Keep critical content inside platform-safe margins.
-
-## Preview, inspect, and repair
-
-Read `references/quality-gates.md`. Complete all relevant gates:
-
-1. factual and editorial;
-2. audio and pronunciation;
-3. timing and caption synchronization;
-4. layout and typography;
-5. motion and continuity;
-6. technical output.
-
-Inspect representative frames at scene starts, middles, transitions, and caption-dense moments. Frame inspection verifies visuals; use waveform or transcript alignment to verify speech synchronization.
-
-Repair issues, regenerate the preview, and repeat inspection. Do not describe a preview as validated until the checks were actually run.
-
-## Render and hand off
-
-Render MP4 only after validation. Confirm duration, resolution, aspect ratio, frame rate, audio stream, and playback.
-
-Deliver:
-
-- final MP4 or the furthest completed artifact;
-- separate publishing cover image;
-- approved narration and evidence table;
-- transcript/captions;
-- storyboard and production manifest;
-- asset and music source record;
-- reusable source/template;
-- concise validation results and any unresolved limitations.
-
-State which steps were automated, which sources were used, and which decisions still require human judgment.
-
-## Source
-
-Rewritten for AICUT from [`yuwenbin121/book-video-production-skill`](https://github.com/yuwenbin121/book-video-production-skill)
-(MIT), SKILL.md. The method is that skill's own; the way it runs is AICUT's.

package/skills/brag/SKILL.md

@@ -1,195 +0,0 @@
----
-name: brag
-description: "Turn the current project website or app into a short, polished, shareable launch video using Hyperframes. Use when someone says \"/brag\", \"use $brag\", \"let's brag about this\", \"make a launch video\", \"turn this into a video\", or wants to share what they built. Reads the project code directly, no live URL or screenshots needed. Runs in AICUT: the film is film.html, a page per shot, drawn by `aicut look` and `aicut render`."
-license: MIT
-source: "mattivilola/brag-codex"
----
-
-# /brag
-
-## Running it in AICUT
-
-AICUT's film is `film.html`: it cuts pages, footage, pictures and sound together on tracks. A page is a web page that
-sets `window.film = { frame(t) }` and draws that moment — no framework, no build step, no composition object, no
-`registerRoot`. Five shots are five pages cut together, not one page that changes scene, so each shot stays trimmable
-and replaceable in Studio.
-
-```sh
-npx -y @autoaiai/aicut@latest          # the manual: the format and the commands, always current
-npx -y @autoaiai/aicut@latest open .   # make this folder a film and print Studio's address
-npx -y @autoaiai/aicut@latest look     # a contact sheet of the whole film, its sound, and a determinism check
-npx -y @autoaiai/aicut@latest look 4.5 # one frame, full size — look at the PNG, not at the code
-npx -y @autoaiai/aicut@latest render   # the MP4, beside film.html
-```
-
-Run the manual before the first film: it is authoritative for the format, and this skill is about the method above it.
-
-Where the words below name another program's part, it becomes this:
-
-| elsewhere | in AICUT |
-|---|---|
-| composition, timeline, project file | `film.html`; tracks and clips inside it |
-| a scene, a sequence, a slide | a page (`pages/<name>.html`) cut onto a track |
-| `useCurrentFrame()`, `interpolate()` | `frame(t)` — the picture depends only on `t`, never on state or `Date.now()` |
-| a still, a screengrab | `aicut look <t>` |
-| `render` / `build` / `export` | `aicut render` (the person can also export from Studio) |
-| a dev server, a preview window | `aicut open` and the Studio address it prints |
-
-Hard rules that hold whatever the method says: the picture depends only on `t`; every value in the person's prompt is
-binding; look at the frames `look` writes before saying a film is done.
-
-
-You built it. Now let's brag about it.
-
-`/brag` turns the current project website or app into a short, polished, shareable launch video using Hyperframes. It is narrow, opinionated, and fun.
-
-## What this skill does
-
-1. Reads the current project code to understand the app.
-2. Plans a short brag concept specific to this project.
-3. Scripts and storyboards the video.
-4. Hands a focused composition brief to Hyperframes.
-5. Validates, renders, and writes share copy.
-
-## Parsing the invocation
-
-The user may invoke with natural language or flags:
-
-```
-/brag
-/brag --tone chaotic
-/brag --tone polished --format vertical
-/brag this. Make it feel like a ridiculous startup launch.
-```
-```
-Parse these options:
-
-| Option | Values | Default |
-|---|---|---|
-| `--tone` | preset or freeform description | inferred |
-| `--format` | `landscape`, `vertical`, `square` | `landscape` |
-| `--duration` | seconds | auto (15-25s) |
-| `--no-music` | flag | music on |
-| `--no-sfx` | flag | sfx on |
-| `--title` | string | inferred from project |
-
-Tone can be a preset (`default`, `polished`, `yc-parody`, `chaotic`, `deadpan`, `cinematic`, `app-store`) or a creative direction such as "fake Series A launch from 2016", "museum exhibit", or "overproduced mobile game ad".
-
-When the user gives freeform tone direction, map it to the nearest preset for pacing and structure, but preserve the user's direction in the plan and composition brief.
-
----
-```
-
-## Output directory
-
-By default, output goes to `brag-output/`. To avoid overwriting previous runs, use a timestamped directory:
-
-```
-brag-output-2026-05-04-143022/
-```
-```
-Use a timestamp when:
-- The user explicitly asks for a new run without overriding previous results
-- A `brag-output/` directory already exists in the project
-
-Generate the timestamp at the start of the run (`YYYY-MM-DD-HHmmss`) and use it consistently for all output paths in that run: plan, brief, composition, render, and share copy.
-
----
-```
-
-## Step 1: Inspect the project
-
-**Read:** [references/step-1-inspect.md](references/step-1-inspect.md)
-
-Scan the project directory and extract the information needed to plan the brag video.
-
-**Gate:** You can answer all 9 questions in the brag planning rubric.
-
----
-
-## Step 2: Plan and storyboard
-
-**Read:** [references/step-2-plan.md](references/step-2-plan.md)
-
-Write `<output-dir>/brag-plan.md` (where `<output-dir>` is `brag-output/` or the timestamped variant chosen above). Answer the planning rubric. Commit to a creative angle. Write the beat-by-beat storyboard including scenes, text, timing, transitions, and SFX cues.
-
-When music is selected, include a compact `Music cue guidance` section: read the bundled track's cue preset from `assets/music/cues/` if present, otherwise note cues will be detected at composition time (any track now supports beat sync — see `references/audio.md`). Cue metadata is optional timing guidance only: story, readability, pacing, and product clarity stay primary.
-
-**Gate:** `<output-dir>/brag-plan.md` exists with a full storyboard. Scene durations sum to 15–25 seconds.
-
----
-
-## Step 3: Hand off to Hyperframes
-
-**Read:** Any available Hyperframes skill, local Hyperframes instructions, or project Hyperframes docs if present.
-**Read:** [references/step-3-compose.md](references/step-3-compose.md)
-**Read:** [references/audio.md](references/audio.md)
-
-Write the composition brief and use Hyperframes to create the video implementation in `<output-dir>/composition/`.
-
-`/brag` owns the product angle, source material, storyboard, tone, format, audio selection, music cue guidance, and delivery expectations. Hyperframes owns the concrete composition structure, exact animation timing, animation mechanics, runtime choices, linting rules, and render workflow.
-
-**Gate:** `npx hyperframes lint` passes with zero errors inside `<output-dir>/composition/`.
-
----
-
-## Step 4: Validate, render, and deliver
-
-**Read:** [references/step-4-deliver.md](references/step-4-deliver.md)
-
-Validate, preview, render to `<output-dir>/brag.mp4`, and write `<output-dir>/share-copy.txt`.
-
-**Gate:** `<output-dir>/brag.mp4` exists. Share copy is written.
-
----
-
-## Tone system
-
-Seven tone presets ship with `/brag`. Each changes scripting energy, pacing, typography personality, and transition style. Presets are defaults, not limits.
-
-Full definitions: [references/tones.md](references/tones.md)
-
-| Tone | Energy | One-liner |
-|---|---|---|
-| `default` | Playful, clean, postable | The good-vibes default |
-| `polished` | Serious, elegant | For projects that are not jokes |
-| `yc-parody` | Deadpan startup energy | Fake seriousness applied to absurd projects |
-| `chaotic` | Fast, loud, aggressive | Over-the-top and unhinged |
-| `deadpan` | Calm, dry, understated | The joke is that nothing is a joke |
-| `cinematic` | Dramatic, trailer-scale | Big motion, bigger claims |
-| `app-store` | Smooth, feature-card clean | Corporate but not boring |
-
-Always allow a freeform creative direction to refine or override the preset.
-
----
-
-## Creative laws
-
-These apply to every brag video regardless of tone.
-
-**Short.** 15–25 seconds. Not one second more without a reason.
-
-**Readable.** Keep the pace high through motion and cuts, never by flashing text. Every line a viewer must read holds long enough to read it (short label ~0.8s settled; a sentence ~0.3s per word). Fast-in, then hold — never fast-in, then gone.
-
-**Specific.** The video must feel like it was made for this exact project, not any project.
-
-**Show the thing.** At least one scene must display actual UI, copy, or a key visual from the product. No abstract filler.
-
-**No generic SaaS language.** "Streamline your workflow" is banned. Use the project's actual copy and claims.
-
-**The hook is everything.** The first 2 seconds determine whether someone keeps watching. Plan the hook before anything else.
-
-**Funny earns its place.** Humor should come from the project's absurdity, not from trying to be funny.
-
-**Pattern:**
-```
-Hook (2-3s) → Reveal (2-4s) → 2-3 sharp highlights (5-12s) → Punchline/outro (2-4s)
-```
-```
-Adapt this. Not every project needs exactly 3 highlights. The pattern is a starting shape, not a template.
-```
-
-## Source
-
-Rewritten for AICUT from [`mattivilola/brag-codex`](https://github.com/mattivilola/brag-codex)
-(MIT), skills/brag/SKILL.md. The method is that skill's own; the way it runs is AICUT's.