Discover and explore top open-source AI tools and projects—updated daily.
Personal Claude agent and command framework
New!
Top 68.5% on SourcePulse
This repository consolidates personal configurations, agents, commands, and output styles for the Claude AI assistant, serving as a practical resource for inspiration and customization. It targets users aiming to build custom agents, manage Claude interactions via slash commands, and enhance prompt engineering with contextual data, offering a glimpse into a personalized Claude workflow and potential extensions.
How It Works
The project structures Claude-related assets into distinct directories such as /agents
, /commands
, and /output-styles
, facilitating modularity and inspiration. Core functionality is driven by custom slash commands, exemplified by /init-workspace
, designed to streamline the installation of agents and MCPs (likely "My Claude Projects") directly into local development environments. This approach aims to simplify the setup and management of custom Claude functionalities. Furthermore, files within /templates
and /docs
are integrated into Claude's system prompts, enabling the AI to access and reference specific information for improved contextual understanding and task execution.
Quick Start & Requirements
Initial setup involves executing custom slash commands like /init-workspace
. The README strongly advises utilizing a separate "mcp boilerplate repo" for simplified creation of new MCPs, which includes commands for local installation. Specific technical prerequisites beyond a functional Claude environment are not detailed, implying a reliance on the existing Claude setup.
Highlighted Details
agents-library
, mcp-library
) for modularity and potential reuse./templates
, /docs
) within system prompts, allowing for dynamic information retrieval.Maintenance & Community
No information provided in the README.
Licensing & Compatibility
No information provided in the README.
Limitations & Caveats
The repository is explicitly described as a "mess" and a personal collection where "not everything is actively used." This suggests potential disorganization, incompleteness, or the presence of outdated elements, positioning it primarily as a source of inspiration rather than a polished, production-ready toolset. Users should expect a high degree of manual configuration and potential debugging.
2 days ago
Inactive