This project provides a collection of independent, open-source Agent Skills for AI content creation, covering images, audio, and video. It targets developers and users seeking flexible AI content generation tools, offering tiered control from fully automated AI partners to direct model-level manipulation. The primary benefit is a modular, multi-model framework that simplifies access to diverse AI generation capabilities.

How It Works

The project is structured around distinct "skills," each addressing a specific AI content creation problem and designed to operate independently. This decoupled architecture allows users to integrate individual skills or compose them as needed without complex orchestration. The "Studios" skills are central, abstracting multiple underlying AI models (e.g., image, video, audio generators) by offering auto-routing to the best provider or explicit model selection, thereby simplifying access to diverse generation capabilities and reducing the need to manage numerous APIs.

Quick Start & Requirements

Installation can be done via agent frameworks like OpenClaw (npx playbooks add skill pexoai/pexo-skills), Claude Code (/plugin marketplace add pexoai/pexo-skills), ClaWHub (clawhub install pexoai/<skill-name>), or by copying skill folders into ~/.cursor/skills/ for Cursor. A single API key is required to unlock all capabilities across providers. No specific OS, Python version, or hardware prerequisites are detailed.

• Website: pexo.ai
• Feedback: GitHub Issues

Highlighted Details

• Multi-Model Abstraction: "Studios" skills consolidate access to numerous leading AI models (e.g., Midjourney, Flux, Ideogram, Kling, Veo, ElevenLabs) under a single interface, simplifying model selection and utilization.
• Decoupled Skill Architecture: Each skill operates independently, eliminating inter-skill dependencies and enabling flexible composition or standalone use.
• Unified API Key Management: A single API key is sufficient to access all integrated services, streamlining credential management across various AI providers.
• Inspectable Open Source: All components, including skills, proxies, and tools, are open source, allowing for transparency and self-hosted deployments.

Maintenance & Community

The project is built and maintained by @pexoai. Further community engagement details like Discord/Slack or a roadmap are not specified in the provided README.

Licensing & Compatibility

The project is licensed under the MIT license, which is permissive for commercial use and integration into closed-source projects.

Limitations & Caveats

The README does not detail specific limitations, unsupported platforms, or known bugs. The modular design implies that complex, multi-skill orchestration might require custom implementation by the user.