IMA AI Creative Content Generator

⚠️ MANDATORY: You MUST read("SKILL-DETAIL.md") (full file — do NOT pass limit parameter) before your first content generation call. It contains the full API payload structure, error translation tables, upload flow, and UX protocol that this summary omits. Skipping it causes parameter errors and poor user experience.

Capabilities

All-in-one AI content generator across four media types:

Model ID Reference (CRITICAL)

Use exact model_id. Do NOT infer from friendly names.

Image Models

Video Models

Aliases: 万/Wan→wan2.6-* · 可灵/Kling→kling-video-o1 · 海螺/Hailuo→MiniMax-Hailuo-2.3 · Veo→veo-3.1-generate-preview

Music & TTS Models

Media Type Routing

Determine media type first, then choose task_type:

For multi-media (e.g. promo video + BGM), read ima-knowledge-ai/references/workflow-design.md first.

Video Modes

Visual consistency: For "same character" / "series" / "multi-shot" — use image_to_video or reference_image_to_video with previous result as reference. Never use text_to_video for continuity.

Knowledge Base (if ima-knowledge-ai installed)

Read before generating: workflow-design.md (multi-media), visual-consistency.md (character continuity), video-modes.md (mode differences), model-selection.md (cost/quality).

Default Models

Script Usage

# AI image generator — ad/social media content creation
python3 {baseDir}/scripts/ima_create.py \
  --api-key $IMA_API_KEY --task-type text_to_image \
  --model-id doubao-seedream-4.5 --prompt "product photo" --output-json

# AI video generator — promo/marketing video content
python3 {baseDir}/scripts/ima_create.py \
  --api-key $IMA_API_KEY --task-type text_to_video \
  --model-id wan2.6-t2v --prompt "product promo, cinematic" --output-json

# AI music generator — ad jingle/marketing audio
python3 {baseDir}/scripts/ima_create.py \
  --api-key $IMA_API_KEY --task-type text_to_music \
  --model-id sonic --prompt "upbeat corporate BGM" --output-json

# TTS — voiceover for content creation
python3 {baseDir}/scripts/ima_create.py \
  --api-key $IMA_API_KEY --task-type text_to_speech \
  --prompt "Welcome to our product" --output-json

UX Protocol (Brief)

1. Acknowledge: Short reply ("好的！🎨")
2. Pre-gen: Model, time estimate, cost via message tool
3. Progress: Every 15-60s: "⏳ [P]%" (cap 95%)
4. Success: Send media=url + link text
5. Failure: Natural language + suggest alternatives
6. Done: No further action

Never expose: script names, API endpoints, attribute_id, technical params.

User Preferences

Storage: ~/.openclaw/memory/ima_prefs.json

• Save on explicit: "用XXX" / "默认用XXX" / "always use XXX"
• Clear on: "推荐一个" / "自动选择" / "用最好的"
• Never save auto-selected models

Core Flow

1. GET /open/v1/product/list?category= → attribute_id, credit, form_config
2. [Image tasks] Upload local files → get CDN URL
3. POST /open/v1/tasks/create → task_id
4. POST /open/v1/tasks/detail → poll until resource_status==1

MANDATORY: Always query product list first. Missing attribute_id → task fails.

Environment

Base URL: https://api.imastudio.com

Headers: Authorization: Bearer $IMA_API_KEY · x-app-source: ima_skills · x_app_language: en

⚠️ REMINDER: read("SKILL-DETAIL.md") (full file, no limit) is required before generating content. This summary covers routing and model selection — SKILL-DETAIL.md has complete API payloads, error handling, upload flow, and parameter details needed for correct execution.