nano-banana-image-ad

これはレビュー時に保存された抜粋です。完全で最新の内容は外部ソースを確認してください。

---
name: nano-banana-image-ad
description: >-
  Generate one or more standalone Meta image-ad creatives via Nano Banana 2 / Nano Banana Pro (Gemini Flash Image family) through the Arcads external API. Locks the model family, auto-strips platform chrome, enforces edge-safe layouts. Use when the user asks for a "Nano Banana ad", "Gemini image ad", "nano-banana-2 ad creative", "make a static image ad with Gemini", or anchors on a need for photoreal / lifestyle / multi-reference / handheld-object / clay-texture ad creatives (sticky-note flatlays, held-whiteboard signs, lifestyle portraits, ingredient collages, OOH photography). Does NOT trigger on ChatGPT Image cues — use chatgpt-image-ad for those.
---

# nano-banana-image-ad (Arcads)

Generate one or more **standalone Meta ad image creatives** via Arcads' `POST /v2/images/generate` with the Nano Banana model family (default `nano-banana-2`). Hands the image paths off to your Meta-ad-builder skill — this skill does not upload to Meta itself.

## Read order

1. **This file** — Arcads-specific endpoint, auth, presigned upload flow, workflow phases.
2. **[shared/skills/nano-banana-image-ad/prompting/guide.md](../../shared/skills/nano-banana-image-ad/prompting/guide.md)** — model-specific prompting (what Nano Banana is good/bad at, when to switch to gpt-image-2).
3. **[shared/skills/image-ad-prompting/prompting/prompt-library.md](../../shared/skills/image-ad-prompting/prompting/prompt-library.md)** — 30+ validated templates with per-model notes.
4. **[shared/skills/image-ad-prompting/prompting/safety-suffixes.md](../../shared/skills/image-ad-prompting/prompting/safety-suffixes.md)** — the 3 always-on guards.
5. **[scripts/generate_image.py](scripts/generate_image.py)** — the helper script (Python stdlib only).

## Hard rules — never relax

1. **Model is in the Nano Banana family.** The script accepts `nano-banana-2` (default), `nano-banana-pro` (Gemini 3 Pro Image, higher cost / locked identity), `nano-banana-edit` (inpaint-focused), or `nano-banana` (legacy). Anything else is refused. If the user asks for gpt-image-2, point them at `chatgpt-image-ad`.
2. **No platform/screenshot chrome in output.** `NO_CHROME_SUFFIX` is always on (override only with `--allow-chrome`).
3. **Edge-safe + glyph-safety suffixes always on** unless `--no-safe-zone` is explicit.
4. **Max 14 reference images.** Hard Arcads cap for Nano Banana. Script enforces.
5. **No Meta upload from this skill.** Image generation only. The user has a separate ad-builder skill in their stack — hand off via filesystem paths.
6. **Always present a credit-cost estimate before generating.** Each Nano Banana call is one image; multiply by `--n`. `nano-banana-pro` costs more than `nano-banana-2` — surface the per-model rate from `logs/arcads-api.jsonl`.

## Prerequisites

- `.env` containing `ARCADS_BASIC_AUTH` (preferred) OR `ARCADS_API_KEY`
- Optional: `PRODUCT_ID`, `PROJECT_ID` in `.env` for session-folder organization
- Reference images on local disk. The script handles the Arcads presigned-upload flow internally.

## Configuration

- **Base URL:** `https://external-api.arcads.ai` (or `ARCADS_BASE_URL`).
- **Auth:** HTTP Basic. The script prefers a pre-encoded `ARCADS_BASIC_AUTH`; falls back to encoding `ARCADS_API_KEY`.
- **Endpoint:** `POST /v2/images/generate`; poll `GET /v1/assets/{id}` until `status: generated`.
- **Reference uploads:** `POST /v1/file-upload/get-presigned-url` returns `{presignedUrl, filePath}`; `PUT` bytes to `presignedUrl`; pass the `filePath` in `referenceImages`. Single-use — re-uploaded fresh per variant by the script.

## Generation modes

| Mode | When to use | Required | Optional |
|---|---|---|---|
| `image` (default) | Brand-new ad image. | `--prompt`, `--aspect-ratio` | `--image-ref` (up to 14) |
| `image_edit` | Modify a `--source` image. | `--prompt`, `--source` | `--image-ref` (up to 14) |

## Supported aspect ratios

`1:1`, `16:9`, `9:16`. **Only these three are accepted by Arcads' `/v2/i