modellix-casting
/modellix-casting
Planuje jeden prompt tablicy obsady na postać: wybiera renderowanie, blokuje tożsamość i format, a potem przekazuje prompt do wykonania.
- Autor
- Modellix ↗
- Repozytorium
- modellix-skill ↗
- Licencja
- MIT
- Ostatnia zmiana repozytorium
- 12.11.2026
Ocena i bezpieczeństwo
- Pliki w paczce
- 10
- Skrypty
- 0
- Komentarze HTML
- 0
- Adresy URL w SKILL.md
- 0
Brak sygnałów. Repozytorium zmienione 12.11.2026, bez odwołań do wycofanych modeli i API.
MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.
Co robi i kiedy
Planuje jeden prompt tablicy obsady na postać: wybiera renderowanie, blokuje tożsamość i format, a potem przekazuje prompt do wykonania.
Wyzwalacze z opisu
Oryginalny opis (EN)
Plan one casting-board prompt per character. Use this skill whenever the user asks for a casting sheet, talent board, 选角图, 定妆照, or a turnaround lookbook — even if they only say "show this character from every angle on one board." Recommend a rendering, lock the identity, confirm the aspect, then hand the prompt to modellix-design. Skip comic model sheets, posters, article covers, and requests that only need an already finished prompt executed.
Jak działa
- Route the request
- Plan the board
- Brief
- Revision
Podgląd SKILL.md
--- name: modellix-casting description: Plan one casting-board prompt per character. Use this skill whenever the user asks for a casting sheet, talent board, 选角图, 定妆照, or a turnaround lookbook — even if they only say "show this character from every angle on one board." Recommend a rendering, lock the identity, confirm the aspect, then hand the prompt to modellix-design. Skip comic model sheets, posters, article covers, and requests that only need an already finished prompt executed. license: MIT compatibility: No external tools required; pairs with the modellix-design skill for image execution. metadata: author: Modellix version: "1.0.0" modellix-hermes-tags: casting,character,turnaround,prompt-engineering,modellix --- # Modellix Casting Turn a character into one prompt for one casting board. This skill plans the image. `modellix-design` generates it. Do not submit, poll, download, authenticate, or inspect model schemas here. Do not run the Modellix CLI. Do not recommend a model, a provider, or a slug. When the user wants the board made, finish the brief first and then hand the prompt to `modellix-design`. A knowledge comic, including the plain character sheet a comic uses between pages, belongs to `modellix-comic`. A poster, product shot, or single portrait belongs to `modellix-image-prompt-engineering`. An article cover belongs to `modellix-cover-image`. A set of illustrations placed through an article belongs to `modellix-article-illustrator`. A card for a named network belongs to `modellix-social-card`. A commerce UGC video belongs to `modellix-ecomm-ugc-video`. Files already on disk belong to `modellix-ffmpeg` or `modellix-moviepy`. ## Route the request Use this skill when the user wants one board that shows the same character as a hero portrait, an expression grid, and a full-body turnaround. Wording includes casting sheet, talent board, 选角图, 定妆照, and turnaround lookbook. Ask once for the character when the user names the format but not the person. Do not invent a name, an age, or a costume to fill an empty request. Skip or keep the pass short when: - The user already supplied a finished casting prompt and asked only to run it. Pass that prompt to `modellix-design`. - The request is a knowledge comic, educational comic, biography comic, or four-panel strip, including the plain sheet that keeps a comic cast consistent. Use `modellix-comic`. - The request is one poster, product shot, or single portrait. Use `modellix-image-prompt-engineering`. - The request is one article, blog, or newsletter cover. Use `modellix-cover-image`. - The request is a set of illustrations placed through an article. Use `modellix-article-illustrator`. - The request names a network and a card, cover, thumbnail, or carousel. Use `modellix-social-card`. - The request is a commerce UGC video that still needs a shot plan. Use `modellix-ecomm-ugc-video`. - The request is video editing, credentials, task status, or a download. ## Plan the board Read `references/layout.md` before describing the frame. Read `references/selection.md` to choose the rendering and the layout example. Read `references/prompt-shape.md` before writing the prompt. 1. Copy only the identity the user gave. Keep age range, face, hair, build, and costume in their words. Drop credentials, tokens, and secrets. Use an original character. Do not name a real celebrity or a copyrighted character. A historical person is described by era, role, and costume, not by a claim that the picture is a photograph of that person. 2. Write one identity lock: age range, face, hair, build, costume, and two or three color names. The same sentences appear in the hero, the expression grid, and the turnaround. 3. Recommend one rendering and the aspect. Give a one-line reason. `studio` is the usual recommendation: a photographic studio, an ivory ground, and even soft light. `illustrated` keeps the same board in a drawing style the user named. `period` keeps the same board and takes the costume from the era the user named. Aspect defaults to 16:9. If the user named no costume, recommend one plain studio outfit and the closest layout example. Do not invent a costume they already contradicted. 4. A choice the user already named is the confirmation. If rendering, aspect, or a missing costume is still open, ask once for the open set and wait. "Use the recommendation" is a confirmation. Do not write the prompt before that. 5. Write one prompt for one board. The image contains no letters, numbers, logos, or watermarks. Describe the left 3×3 headshots, the center portrait, and the right full-body front, three-quarter, and back views. 6. When the user asks for a separate front, side, or back file, add one prompt per named view and keep the board. Do not split the board unless they ask. Each extra view is a separate generation later. 7. One character is one board. Plan at most four boards unless the user names a higher count. When the user supplies a picture of the person, name that path with usage `direct`. For `studio`, also name one layout example from `references/examples/` with usage `layout`. Pick the example whose costume is closest; otherwise use `references/examples/studio-board-suit.webp`. The prompt says the layout example controls the board and the studio light, and does not control face, hair, age, build, or costume. For `illustrated` and `period`, do not attach a photographic example. This skill does not upload files. `modellix-design` decides which schema field receives each path. Until the open choices are confirmed, return the recommendation and the identity you will keep, and leave the prompt empty. ## Brief Return this brief: ```text Casting brief - Character: - Rendering: - Aspect: - Identity lock: - References: none | path / direct or layout - Board prompt: - Extra views: none | view / prompt - Next step: stop after prompt | pass the prompt to modellix-design - Model: <only a slug the user already named | omitted> ``` Leave `Model` blank unless the user already named a slug. Do not browse the catalog from this skill. If the user asked only for a prompt, stop after the brief. If the original request asked for the image, pass the board prompt to `modellix-design`, then any extra views. That skill asks for a model and owns schema validation, credentials, paid submission, waiting, and download. Each file is a separate image. ## Revision When a generated board misses, change the prompt and return a new brief: - The face or costume drifted: restate the identity lock in every zone. - The layout example's person leaked in: say again that the example controls only the board and the light. - An extra person, a label, or a number appeared: restate the 3×3 grid, the three full-body views, and that the image contains no text. - Do not paint corrected type onto the bitmap. `modellix-design` decides whether an existing task can be recovered or a new submission is appropriate.
Pliki w paczce (10)
- SKILL.md 7 KB
- skill.json 1 KB
- evals/evals.json 5 KB
- references/layout.md 2 KB
- references/prompt-shape.md 2 KB
- references/selection.md 1 KB
- references/examples/studio-board-blazer.webp 100 KB
- references/examples/studio-board-knit.webp 96 KB
- references/examples/studio-board-shirt.webp 79 KB
- references/examples/studio-board-suit.webp 88 KB
Wersje
W bazie nie ma innych skilli o nazwie modellix-casting.
Migawka bazy z 13.11.2026 · wersja paczki 467fbdb4371b (skrót SHA-256 całej paczki)
Pobrania w ostatnich 30 dniach: 0 · ranking
Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.
Podobne skille
- 100modellix-comic
Planuje prompty do komiksu edukacyjnego, wiedzowego, biograficznego lub czterokadrowego: poleca styl, ton i układ, potwierdza liczbę stron, proporcje i język, ustala obsadę i przekazuje każdą stronę do modellix-design. - 100modellix-cover-image
Planuje jeden prompt do okładki artykułu, bloga lub newslettera: poleca typ, paletę, renderowanie, ilość tekstu, nastrój i krój, potwierdza je, cytuje tytuł źródła i przekazuje generowanie do modellix-design. - 100creator-character-designer
Projektuje postacie jako spójną całość: funkcja dramaturgiczna, psychologia, biografia, tożsamość wizualna, kostium, rekwizyty, profil castingowy i ekspresje kodowane FACS, dla filmu tworzonego z AI. - 100character-design
Prowadzi nowicjusza od pomysłu do zatwierdzonych arkuszy postaci dla comicblock, pytając krok po kroku. - 90character-sheet
Tworzy i blokuje spójną postać w NovaField: projekt z opisu lub zdjęć, arkusz, zdjęcia testowe i zapis wybranych ujęć jako zasobu obsady. - 100voice-act
Renderuje kwestie postaci przez ElevenLabs i tworzy pliki MP3 do rozdziałów, w czterech trybach, od przesłuchań po lektorat.