Communitygithub.com

Media Tools (Free Image Generation)

Free image reading & generation for DeepSeek Harness — paste an image into any chat, even text-only sessions. Supports 9 languages, no API key required for storage.

What is Media Tools (Free Image Generation)?

Media Tools is a skill for AI agent harnesses (like DeepSeek Harness) that enables free image generation and reading capabilities. It allows you to paste an image into any chat session, including text-only ones, and generate new images using the SiliconFlow Kolors model without ever storing the API key inside the skill itself. The skill provides a simple CLI command to generate images with customizable dimensions, and it respects rate limits (approximately 2 images per minute). It is designed to protect user privacy by only writing generated assets to the specified output path and never syncing them to public directories. The skill supports multiple languages and requires only a free SiliconFlow API key, which can be set via environment variables or secret files.

Works with~Claude Code✓Codex CLI~Cursor
npx skills add akqwpeter-prog/dsh-media-skills
View original→Browse all skills

Ask in your favorite AI

Open a new chat with this agent skill pre-loaded.

OpenAIChatGPTClaudeClaudeGeminiGeminiGrokGrokPerplexityPerplexityDeepSeekDeepSeek

Documentation

What does Media Tools (Free Image Generation) do?

Free image generation based on SiliconFlow Kolors. The API key is never stored in this skill.

Generate an Image

python3 scripts/generate.py "<prompt>" <output path.png> [size=1024x1024]
  • Free, no watermark, rate limited to approximately 2 images per minute.
  • If a model id returns "Model disabled", use the same key to list available models via GET https://api.siliconflow.cn/v1/models.

Key

  • SILICONFLOW_API_KEY (SiliconFlow, free model Kwai-Kolors/Kolors). How to obtain: Register/Log in at siliconflow.cn → "API Keys" → Create a new key and copy it (Kolors is free).
  • Reads from environment variable first; otherwise reads from ~/.dsh/secrets/media-tools.env or ~/.codex/secrets/media-tools.env (each line KEY=value, permissions 600).
  • Never commit the key to a repository, write it into the skill, or paste it into a public file.

Privacy

  • Generated assets are only written to the specified output path, never automatically synced to a public directory.
  • Do not reproduce sensitive reference images or personal information in generated results without permission.

Related Skills

steipete/sag

ElevenLabs text-to-speech with mac-style say UX.

community

steipete/oracle

Oracle CLI second-model review/debug/refactor/design with selected files, dry-run token checks, API or browser engine.

community

steipete/peekaboo

Capture and automate macOS UI with the Peekaboo CLI.

community

mattpocock/design-an-interface

Generate multiple radically different interface designs for a module using parallel sub-agents. Use when user wants to design an API, explore interface options, compare module shapes, or mentions "design it twice".

community

mattpocock/prototype

Build a throwaway prototype to flesh out a design — a runnable terminal app for state/business-logic questions, or several radically different UI variations toggleable from one route.

community

mattpocock/grill-me

Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".

community