IMA AI Creative Content Generator — Image, Video, Music, Ad, Marketing, Social Media Content Creation

All-in-one AI content generator for creative, marketing, and social media content creation. AI image generator with SeeDream 4.5, Midjourney, Nano Banana 2/P...

Security Scan

The skill's code, runtime instructions, and requested environment access are internally consistent with an AI content-generation tool and do not request unrelated credentials or suspicious installs.

Purpose & Capability

Name/description (multimodal AI content generation) align with required items: python3, requests, and a single API key (IMA_API_KEY). The script talks to imastudio endpoints and upload services that match the described functionality (product list, task create, local-image upload).

Instruction Scope

SKILL.md requires reading SKILL-DETAIL.md before first use and optionally reading ima-knowledge-ai references when installed. These reads are declared in instructionScope and are consistent with the skill’s need for product/model tables and workflow guidance. The mandatory 'read full SKILL-DETAIL.md' is a usability requirement (not a secret-exfiltration directive) but does grant the skill permission to read its own included documentation fully.

Install Mechanism

No install spec is provided; the skill is instruction + included Python scripts. Required runtime (python3) and package (requests) are reasonable and proportionate. No downloads from untrusted URLs or archive extraction occur during install.

Credentials

Only one credential (IMA_API_KEY) is required and declared as the primary credential. The code uses that key for api.imastudio.com and (when uploading local images) imapi.liveme.com, which the docs explicitly state. No unrelated secrets or broad cloud credentials are requested.

Persistence & Privilege

The skill writes per-user preferences and logs under ~/.openclaw (declared in persistence). It does not require always:true and does not modify other skills. Logs may contain operational details (but do not appear to store the API key). The skill may optionally read ima-knowledge-ai reference files if that skill is installed (this cross-skill read is declared and optional).

Assessment

This skill appears to do what it says: it runs Python scripts that call api.imastudio.com and (for local file uploads) imapi.liveme.com and then uploads binaries to presigned CDN hosts. Before installing or using it: 1) Only provide a dedicated IMA_API_KEY with limited scope (so you can revoke/regenerate it easily if needed). 2) Review SKILL-DETAIL.md and SECURITY.md (the skill requires you to read them) to confirm model IDs, upload behavior, and cost rules. 3) Be aware that local images are uploaded to a CDN via presigned URLs (the API key is sent to imapi.liveme.com during the upload-token step as documented). 4) Logs and a small prefs file are stored under ~/.openclaw — if that is sensitive, clear or relocate them. 5) If you do not trust api.imastudio.com or the imapi.liveme.com upload service, do not supply your API key. 6) Use network monitoring or a scoped API key on first run if you want to verify exact network behavior.

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:

Media	Models	Content Creation Use Cases
Image	SeeDream 4.5, Midjourney, Nano Banana 2/Pro	Ad creatives, social media graphics, product photos, posters
Video	Wan 2.6, Kling O1/2.6, Veo 3.1, Sora 2 Pro, Hailuo 2.3/2.0, Pixverse, SeeDance, Vidu, Ima Sevio	Promo videos, short videos, marketing clips, product demos
Music	Suno sonic v5, DouBao BGM/Song	Ad jingles, background music, social media audio
TTS	seed-tts-2.0	Voiceovers, narration, ad audio

Model ID Reference (CRITICAL)

Use exact model_id. Do NOT infer from friendly names.

Image Models

Name	model_id	Cost
SeeDream 4.5 🌟	doubao-seedream-4.5	5 pts
Midjourney 🎨	midjourney	8-10 pts
Nano Banana2 💚	gemini-3.1-flash-image	4-13 pts
Nano Banana Pro	gemini-3-pro-image	10-18 pts

Video Models

Name	model_id (t2v / i2v)	Cost
Wan 2.6 🔥	wan2.6-t2v / wan2.6-i2v	25-120 pts
Kling O1	kling-video-o1	48-120 pts
Kling 2.6	kling-v2-6	80+ pts
Hailuo 2.3	MiniMax-Hailuo-2.3	32+ pts
Hailuo 2.0	MiniMax-Hailuo-02	5+ pts
Veo 3.1	veo-3.1-generate-preview	70-330 pts
Sora 2 Pro	sora-2-pro	122+ pts
Pixverse V5.5	pixverse	12-48 pts
SeeDance 1.5 Pro	doubao-seedance-1.5-pro	20+ pts
Vidu Q2	viduq2 / viduq2-pro	5-70 pts
IMA Video Pro	ima-pro	varies
IMA Video Pro Fast	ima-pro-fast	varies

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

Name	model_id	Cost
Suno sonic v5	sonic	25 pts
DouBao BGM	GenBGM	30 pts
DouBao Song	GenSong	30 pts
seed-tts-2.0	seed-tts-2.0	query product list

Media Type Routing

Determine media type first, then choose task_type:

User keywords	Type	task_type
画/图/image/poster/thumbnail/ad graphic	image	text_to_image, image_to_image
视频/video/promo/short/宣传片/clip	video	text_to_video, image_to_video, first_last_frame_to_video, reference_image_to_video
音乐/BGM/music/jingle/歌	music	text_to_music
语音/TTS/voiceover/narration/配音	speech	text_to_speech

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

Video Modes

Intent	task_type
Text only → video	text_to_video
Image as first frame	image_to_video
Image as style reference	reference_image_to_video
Two images (start+end)	first_last_frame_to_video

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

Task	Default	model_id	Cost
text_to_image	SeeDream 4.5	doubao-seedream-4.5	5 pts
text_to_video	Wan 2.6	wan2.6-t2v	25 pts
image_to_video	Wan 2.6	wan2.6-i2v	25 pts
first_last_frame	Kling O1	kling-video-o1	48 pts
text_to_music	Suno	sonic	25 pts

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=<task_type> → 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.