BACK TO PROJECTS
AI / Agentic

Agentic Skills (MCP)

A collection of Model Context Protocol (MCP) skills for AI agents. Enabling Claude Code and other agents to interact with Cardano, Algorand, and Zama.

Agentic Skills

I am building the interface layer between AI Agents and the physical/digital world. By developing custom skills for the Model Context Protocol (MCP), I enable agents to perform complex blockchain operations and data analysis.

Key Skills

  • Cardano Skill: Stake, transfer, and query UTXOs from Claude Code.
  • Algorand Skill: Interact with Teal contracts and manage ASA (Algorand Standard Assets).
  • Zama Skill: Execute private computations on the fhEVM via AI commands.
  • Charli3 Skill: Fetch real-time oracle data into agent context.

Why it matters

AI agents are limited by their context. MCP skills provide the "hands" that allow them to actually do things on-chain rather than just talk about them.

Tech

  • Framework: Model Context Protocol (MCP)
  • Languages: TypeScript, Python, Shell
  • Integration: Claude Code, Cursor, Windsurf