Skip to main content

carousel-builder

Build branded Instagram carousels end-to-end: Higgsfield generates a theme-locked cover + middle images, character budgets keep copy on-template, and Canva assembles the deck.

Informations de source

Dépôt
authrain-cloud-abdullahformuli/carousel-builder
Dernière activité de la source
1 octobre 2026 à 03:19
Langue détectée de SKILL.md
anglais
Étoiles
0
Forks
0

Options d'installation

Le prompt qui vérifie d'abord la source est sélectionné par défaut. Vous pouvez passer à une commande directe ou télécharger une copie locale.

Vérifiez les fichiers source

Lisez SKILL.md et les fichiers associés affichés par SkillsMP avant de décider de l'installer.

Affichage de SKILL.md

SKILL.md
Instructions source · Aperçu en lecture seule
name
carousel-builder
description
Build branded Instagram carousels end-to-end: Higgsfield generates a theme-locked cover + middle images, character budgets keep copy on-template, and Canva assembles the deck.
# Carousel Builder Build an Instagram carousel end-to-end: 1. Generate slide-1 cover image with **Higgsfield MCP** (4:5). 2. Write all slide text + per-slide image prompts into a markdown file with **locked character budgets**. 3. Generate **theme-matched 16:9 images** for the middle slides (passing the chosen cover as reference). 4. Assemble the deck by duplicating a fixed **Canva template** and replacing text spans and image fills. 5. Finish by running `/short-form-caption` for the post copy. ## Requirements - **Higgsfield MCP**: `nano_banana_2` model for cover and pillar images. - **Canva MCP**: `https://mcp.canva.com/mcp` for templated deck assembly. - **Canva Template**: 8-page carousel template. ## Character Budgets (Hard Caps) - **Pillar headline**: 45–58 characters (including `The [Role]:`, exactly 1 highlighted word). - **Pillar body**: 65–80 characters (ends with 1 word in ALL CAPS). - **Cover headline**: ≤ 22 characters. - **Cover subtext**: 3–4 short italic lines. - **Caption**: ≤ 500 characters total with exactly 5 hashtags. ## Workflow Execution 1. Check Canva template ID and brand context configuration. 2. Interview the topic into 4–6 pillars and CTA keyword. 3. Generate 3–4 4:5 cover options with Higgsfield; user selects the anchor. 4. Draft copy and validate character counts before image generation. 5. Generate 16:9 pillar images using the chosen cover as style reference. 6. Assemble in Canva via duplicate and `update_fill`. 7. Generate copy-paste caption with short-form-caption skill.
Voir sur GitHub