skillZs
★ LIVE SKILL TAGS ★
>>> LIVE SKILLS INDEX <<<
* OPEN SOURCE *
NO LOGIN, NO TRACKING
※ REAL INSTALL DATA ※
← back to all skills
tinyshipcn/tinyship-skills94 installs

tinyship-ai

Configure AI features in TinyShip: chat (OpenAI, DeepSeek, Qwen), image generation (fal.ai, Gemini, Qwen), and video generation. Includes credit cost setup. Use when the user asks to "configure AI", "add AI chat", "set up image generation", "video generation", "AI provider", "add AI feature", or "configure AI credits".

How do I install this agent skill?

npx skills add https://github.com/tinyshipcn/tinyship-skills --skill tinyship-ai
view source ↗

Is this agent skill safe to install?

  • Gen Agent Trust Hubpass

    The skill provides legitimate documentation and instructions for configuring AI features (chat, image, and video generation) within the TinyShip application. It follows security best practices by using environment variables for API keys with placeholders and referencing official provider documentation.

  • Socketpass

    No alerts

  • Snykpass

    Risk: LOW · No issues

What does this agent skill do?

TinyShip AI Integration

Guide for configuring AI capabilities in TinyShip. Built on Vercel AI SDK.

Available Capabilities

CapabilityProvidersConfig File
AI ChatQwen, DeepSeek, OpenAIconfig/ai.ts
Image Generationfal.ai, Gemini, Qwenconfig/aiImage.ts
Video GenerationQwen (DashScope), Volcengineconfig/aiVideo.ts

Ask the user which capabilities they want to enable.

AI Chat Setup

Step 1: Choose Provider

ProviderStrengthsCostAPI Key Source
Qwen (recommended)Chinese/English, cost-effectiveLowAliyun DashScope
DeepSeekStrong coding, very cheapVery LowDeepSeek Platform
OpenAIBest overall performanceHighOpenAI Platform

Step 2: Configure Environment

Add the chosen provider's API key to .env:

# Default provider selection
AI_PROVIDER="qwen"  # or "deepseek" or "openai"

# Qwen
QWEN_API_KEY="sk-xxxxx"
QWEN_BASE_URL="https://dashscope.aliyuncs.com/compatible-mode/v1"

# DeepSeek
DEEPSEEK_API_KEY="sk-xxxxx"

# OpenAI
OPENAI_API_KEY="sk-xxxxx"
OPENAI_BASE_URL=""  # optional: custom gateway URL

Step 3: Verify

  1. Start the dev server
  2. Navigate to /ai
  3. Send a test message
  4. Verify streaming response works

Image Generation Setup

Add the chosen provider's API key to .env:

# fal.ai (recommended, supports many models)
FAL_API_KEY="xxxxx"

# Google Gemini
GOOGLE_GENERATIVE_AI_API_KEY="xxxxx"
GOOGLE_GENERATIVE_AI_BASE_URL=""  # optional: custom gateway

# Qwen (reuses QWEN_API_KEY from chat setup)

Configure available models in config/aiImage.ts. The config file defines which models are available and their default parameters.

Verify at /image-generate.

Video Generation Setup

Add the chosen provider's API key to .env:

# Qwen/DashScope (reuses QWEN_API_KEY)
# No additional config needed if Qwen chat is already set up

# Volcengine
VOLCENGINE_ACCESS_KEY_ID="xxxxx"
# VOLCENGINE_BASE_URL="https://ark.cn-beijing.volces.com"  # optional

Configure available models in config/aiVideo.ts.

Verify at /video-generate.

Credit Costs (Optional)

If you want AI features to consume credits, configure config/credits.ts:

credits: {
  consumptionMode: 'dynamic',
  dynamicChatCostPerKiloToken: 1,
  modelMultipliers: {
    'qwen-turbo': 1.0,
    'qwen-max': 2.0,
    'deepseek-chat': 0.5,
    'gpt-4': 3.0,
    'default': 1.0
  },
  imageGenerationCost: 5,   // credits per image
  videoGenerationCost: 20,  // credits per video
}

Make sure credit purchase plans are set up (use tinyship-payment skill).

Adding a New AI Provider

Read libs/ai/AGENTS.md for the full pattern. Summary:

  1. Install the SDK: pnpm add @ai-sdk/new-provider
  2. Update types in libs/ai/types.ts
  3. Add env key mapping in libs/ai/config.ts
  4. Implement provider in libs/ai/providers.ts
  5. Add env vars to env.example

Reference Files in TinyShip Repo

  • config/ai.ts — chat provider configuration
  • config/aiImage.ts — image generation configuration
  • config/aiVideo.ts — video generation configuration
  • config/credits.ts — credit costs for AI features
  • libs/ai/AGENTS.md — AI library architecture and provider patterns
  • libs/credits/AGENTS.md — credit system architecture
  • docs/user-guide/ai/chat.md — AI chat documentation
  • docs/user-guide/ai/image.md — image generation documentation
  • docs/user-guide/ai/video.md — video generation documentation

Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.

<a href="https://skillzs.dev/skills/tinyshipcn/tinyship-skills/tinyship-ai">View tinyship-ai on skillZs</a>