agentleFS
joebenscoter86joebenscoter86/higgsfield-ugc-workflow/skills/ugc-storyboard-sheet/SKILL.md

ugc-storyboard-sheet

Skill4 starsChanged 5 months agoMITView on GitHub

Step 3 of the ugc-ad pipeline. Generate ONE 16:9 storyboard sheet holding three side-by-side 9:16 panels (Hook / Setup-Action / Recommendation) at framings Tight / Macro / Wide, using the product and character images as references to lock identity. This sheet is the visual narrative map the video model follows. Brand-agnostic.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Generate the sheet: Call generate_image:

    line 20

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

ugc-storyboard-sheet

Produces the three-panel storyboard that dictates the ad's scene sequence and locks visual identity.

Inputs

  • PRODUCT_MEDIA_ID (from ugc-product-profile)
  • CHARACTER_JOB_ID (from ugc-base-character)
  • PROFILE (path to product-profile.md, for the usage steps)
  • DEST (path)

Steps

  1. Confirm the image model + reference roles. Default nano_banana_pro (handles multiple reference images + clean layout/text; 16:9 supported, no max on image references). Call models_explore(action="get", model_id="nano_banana_pro") to confirm. NOTE (validated 2026-06-22): requesting nano_banana_pro may report back as model nano_banana_2 in the job response — this is a server-side alias; references still bind and the 3-panel sheet renders correctly. Not an error.
  2. Generate the sheet. Call generate_image:
    • model: "nano_banana_pro"
    • aspect_ratio: "16:9", count: 1
    • medias: pass the product and character as references, e.g. [{ "value": "<PRODUCT_MEDIA_ID>", "role": "image" }, { "value": "<CHARACTER_JOB_ID>", "role": "image" }] (use the exact role names returned by models_explore).
    • prompt: one horizontal sheet, three equal vertical 9:16 panels side by side, same creator and same product in every panel:
      • Panel 1 — Hook (Tight): tight selfie-framing of the creator reacting/speaking to camera.
      • Panel 2 — Setup/Action (Macro): macro on the creator's hands performing the product's exact usage step (pull from PROFILE).
      • Panel 3 — Recommendation (Wide): waist-up wide of the creator holding/presenting the product to camera.
    • Poll job_status; capture job_id + URL.
  3. Download: curl -fsSL <url> -o "$DEST/storyboard.png".

Output

DEST/storyboard.png + its job_id (report it — the video step uses it as the scene-sequence reference).

Constraints

  • Same face + same product packaging in all three panels (that is the whole point — continuity lock).
  • Three distinct framings (Tight / Macro / Wide) for dynamic progression.
  • 16:9 canvas, three 9:16 panels. Verify the output actually shows three panels before reporting success.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.