Novita AI Multimodal

Execute multimodal tasks using Novita AI: text-to-image, image-to-image, text-to-video, image-to-video, TTS, STT. Use for: generating images, generating vide...

Security Scan

The skill's instructions look like a genuine Novita API integration, but the registry metadata omits the API key/config requirements and the runtime instructions ask the agent to extract keys from user messages and a specific config path—these mismatches and the key-extraction behavior warrant caution.

Purpose & Capability

The skill claims no required env vars/config paths/primary credential in registry metadata, yet the SKILL.md clearly expects an API key (via ~/.novita/config.json, NOVITA_API_KEY, or in-message 'sk_' token). The need for an API key is reasonable for a multimodal API integration, but the metadata omission is an inconsistency that could mislead users about what secrets the skill uses.

Instruction Scope

Runtime instructions explicitly direct the agent to: parse user messages for API keys (tokens starting with 'sk_'), read ~/.novita/config.json, or read NOVITA_API_KEY. Asking the agent to extract API keys from free-form user messages grants broad discretion and increases risk of accidental key exposure. Aside from that, the API endpoints and curl examples stay within the stated multimodal purpose.

Install Mechanism

Instruction-only skill with no install spec and no code files; nothing is written to disk by an installer. This is the lowest-risk install mechanism.

Credentials

The skill legitimately requires a single Novita API key to call external endpoints, which is proportional. However the registry declares no primary credential or required env/config path while the SKILL.md requires ~/.novita/config.json or NOVITA_API_KEY or an in-message key—this metadata mismatch is problematic and reduces transparency. Also the SKILL.md uses $API_KEY in examples while recommending NOVITA_API_KEY in setup, which is ambiguous.

Persistence & Privilege

The skill does not request always: true and does not declare system-wide modifications. It can be invoked autonomously (platform default), which is normal; no elevated persistence or cross-skill config changes are requested.

What to consider before installing

This skill appears to be a straightforward Novita AI integration, but the registry metadata omits the fact that it needs an API key. Before installing or using it: (1) do not paste your private API keys into chat messages—prefer setting NOVITA_API_KEY in your environment or creating the ~/.novita/config.json file; (2) verify you trust the skill source (homepage/source unknown) because the skill will send your API key to https://api.novita.ai to generate media; (3) be aware the SKILL.md asks the agent to look for keys inside user messages (tokens starting with 'sk_')—avoid sending keys in free text to prevent accidental leakage; and (4) expect minor ambiguity in variable names ($API_KEY vs NOVITA_API_KEY/~/.novita/config.json) — ensure the agent or your environment maps them correctly. If you need higher assurance, ask the author to update the registry metadata to declare the required credential and clarify the env/config variable names.

Novita AI Multimodal Execution

Configuration (choose one, by priority)

Method 1: Config File (Recommended)

Create file ~/.novita/config.json:

{
  "api_key": "YOUR_API_KEY"
}

One command setup:

mkdir -p ~/.novita && echo '{"api_key": "YOUR_API_KEY"}' > ~/.novita/config.json

Method 2: Environment Variable

export NOVITA_API_KEY="YOUR_API_KEY"

Method 3: Direct Parameter

Include in request: Please use API Key sk_xxx to generate an image...

---

API Key Reading Logic

1. Check if user message contains API Key (starts with sk_)
2. Check config file ~/.novita/config.json
3. Check environment variable NOVITA_API_KEY
4. None found → Return configuration guide

Configuration guide (only shown when not configured):

You have not configured your Novita AI API Key.

Quick setup (copy and run):
mkdir -p ~/.novita && echo '{"api_key": "YOUR_KEY"}' > ~/.novita/config.json

Get Key: https://novita.ai/settings/key-management

---

Execution Flow (Important!)

User request → Identify task → Get Key → ⚠️ Send prompt first → Execute task → Return result

⚠️ Must Send Progress Prompt First

Before calling the API, you must reply to the user with a message:

🎨 Got it! Generating your image...

Task type: Text-to-Image
Model: Seedream 5.0 Lite
Estimated time: 5-15 seconds
Estimated cost: ~$0.035

Please wait, will send as soon as it's ready ⏳

This message must be sent BEFORE executing the API call! This way users know the task is being processed and won't think the system is stuck.

Progress Templates for Different Tasks

Text-to-Image:

🎨 Got it! Generating your image...
Model: Seedream 5.0 Lite
Estimated time: 5-15 seconds

Text-to-Video:

🎬 Got it! Generating your video...
Model: Vidu Q3 Pro
Estimated time: 1-3 minutes (video generation is slower, please be patient)

TTS:

🔊 Got it! Generating your audio...
Model: MiniMax Speech 2.8 Turbo
Estimated time: 5-15 seconds

Completion Response

✅ Generation complete!

[Image/Video/Audio URL]

Actual cost: $0.035

Video Task Polling Updates

Video generation requires polling, update status every 15 seconds:

🎬 Video generating...
Current status: Processing
Elapsed: 30 seconds
Estimated remaining: 1-2 minutes

---

API Configuration

Setting	Value
Base URL	https://api.novita.ai
Auth	Authorization: Bearer <API_KEY>
Get Key	https://novita.ai/settings/key-management

Task Types and Endpoints

Task	Endpoint	Model
Text-to-Image	/v3/seedream-5.0-lite	Seedream 5.0 Lite
Image Editing	/v3/seedream-5.0-lite	Seedream 5.0 Lite
Text-to-Video	/v3/async/vidu-q3-pro-t2v	Vidu Q3 Pro
Image-to-Video	/v3/async/vidu-q3-pro-i2v	Vidu Q3 Pro
TTS	/v3/async/minimax-speech-2.8-turbo	MiniMax Speech 2.8
STT	/v3/glm-asr	GLM ASR
Task Query	/v3/async/task-result?task_id=xxx	-

---

Execution Templates

Text-to-Image

curl -X POST "https://api.novita.ai/v3/seedream-5.0-lite" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "description"}'

Image Editing

curl -X POST "https://api.novita.ai/v3/seedream-5.0-lite" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "edit instruction", "reference_images": ["image_url"]}'

Text-to-Video

curl -X POST "https://api.novita.ai/v3/async/vidu-q3-pro-t2v" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "description", "duration": 4}'

Image-to-Video

curl -X POST "https://api.novita.ai/v3/async/vidu-q3-pro-i2v" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "motion description", "images": ["image_url"]}'

TTS

curl -X POST "https://api.novita.ai/v3/async/minimax-speech-2.8-turbo" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "text to convert",
    "voice_setting": {"voice_id": "male-qn-qingse", "speed": 1.0},
    "audio_setting": {"format": "mp3"}
  }'

Available voices:

• Male: male-qn-qingse, male-qn-jingying
• Female: female-shaonv, female-yujie

STT

curl -X POST "https://api.novita.ai/v3/glm-asr" \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"file": "audio_url_or_base64"}'

Task Result Query

curl "https://api.novita.ai/v3/async/task-result?task_id=$TASK_ID" \
  -H "Authorization: Bearer $API_KEY"

Status: TASK_STATUS_QUEUED → TASK_STATUS_PROCESSING → TASK_STATUS_SUCCEED

---

Error Handling

Code	Meaning	Action
401	Invalid Key	Check configuration
402	Insufficient balance	Top up at https://novita.ai/billing
429	Rate limited	Wait and retry

Pricing

https://novita.ai/pricing