@MatheusCampagnolo
FreeLocal, agent-first control plane for catalog discovery and skill selection
FreeFree tier
About @MatheusCampagnolo
AAS Core is a local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning. It is backed by over 1,987 agentic skills and includes a CLI, local MCP, and catalog features. Built with Python, it is designed to help agents autonomously discover, validate, and select skills from a vast catalog.
Key Features
Agent-first control plane for local execution
Complete catalog discovery of agentic skills
Agent-owned skill selection and planning
Stack validation and dependency checking
Command-line interface (CLI) for interaction
Local Model Context Protocol (MCP) support
Backed by 1,987+ agentic skills in the catalog
Pros & Cons
Pros
- Open source and freely available on GitHub
- Comprehensive catalog of nearly 2,000 agentic skills
- Includes both CLI and local MCP for flexible integration
- Agent-first design gives autonomy to agents for skill selection
Cons
- May require understanding of agentic concepts and MCP
- Documentation and setup guides may be limited beyond GitHub README
- Primarily targeted at developers and advanced users
Best For
Enabling autonomous agents to discover and select skills from a large catalogValidating software stacks for agentic workflowsPlanning multi-step tasks with agent-owned decision makingLocal development and testing of agent-based systems