Does Agentica support Model Context Protocol (MCP)?

Yes, Agentica has native MCP integration, allowing you to connect agents to MCP servers via configuration files for expanded tool capabilities.

Can I track token usage with this skill?

Yes, the SDK provides 'last_usage()' and 'total_usage()' methods to monitor input, output, and cumulative token processing for cost management.

How do I handle persistent memory in Agentica?

For @agentic functions, you can set 'persist=True' to remember conversation history. For agents created via spawn(), state is automatically maintained across calls to the same instance.

What is Agentica SDK?

Agentica is a Python framework designed for building AI agents that can maintain state, implement functions, use external tools, and coordinate with other agents.

Which LLM models can I use with Agentica?

Agentica supports major models from Anthropic (Claude), OpenAI (GPT-4/5), and any provider available through OpenRouter slugs like Google Gemini.

Agentica SDK

Name: Agentica SDK
Author: scooter-lacroix

byscooter-lacroix

Ciencia de Datos y ML

Builds and orchestrates stateful Python-based AI agents with MCP integration and multi-agent patterns.

Acerca de

The Agentica SDK skill provides a comprehensive framework for developing sophisticated AI agents in Python. It enables developers to transform standard functions into AI-powered tools using the @agentic decorator, spawn autonomous agents with specific premises, and manage complex multi-agent orchestrations. With built-in support for the Model Context Protocol (MCP), token usage tracking, conversation persistence, and OpenTelemetry tracing, this skill is essential for building production-grade AI applications that require structured outputs, tool-use, and robust state management.

Características Principales

Seamless Model Context Protocol (MCP) integration for connecting external tools
Asynchronous agent spawning with custom system prompts and model selection
Support for multi-agent coordination and persistent state across conversation history
Advanced token management including usage tracking and fine-grained limits
@agentic decorator for creating AI-powered Python functions with minimal boilerplate
0 GitHub stars

Casos de Uso

Developing complex multi-agent workflows for automated research or data analysis
Integrating external APIs and legacy tools into LLM-driven applications via MCP
Building stateful chatbots and assistants that maintain context across multiple interactions

Acerca de

Características Principales

Seamless Model Context Protocol (MCP) integration for connecting external tools
Asynchronous agent spawning with custom system prompts and model selection
Support for multi-agent coordination and persistent state across conversation history
Advanced token management including usage tracking and fine-grained limits
@agentic decorator for creating AI-powered Python functions with minimal boilerplate
0 GitHub stars

Casos de Uso

Developing complex multi-agent workflows for automated research or data analysis
Integrating external APIs and legacy tools into LLM-driven applications via MCP
Building stateful chatbots and assistants that maintain context across multiple interactions