Agentica Sdk

The Agentica SDK gives you a Python framework for building AI agents with the @agentic decorator or spawn() function. You get typed return values, conversation persistence, tool passing through scope parameters, and built-in MCP integration for external tools. The decorator pattern is clean for turning functions into agents, while spawn() gives you more control over premises and system prompts. Token tracking and streaming are first-class features, and the error handling is granular enough to catch rate limits versus content filtering. It's opinionated about logging (stdout plus files by default) but you can configure it per-agent or globally. If you're building agents that need to coordinate with each other or wrap existing Python SDKs with agentic behavior, this handles the orchestration layer without forcing you into a heavyweight framework.

skills.sh installs: 474
GitHub stars: 3.9K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add parcadei/continuous-claude-v3 --skill agentica-sdk --agent claude-code

Installs into .claude/skills of the current project.

Files

Agentica SDK Reference (v0.3.1)

Build AI agents in Python using the Agentica framework. Agents can implement functions, maintain state, use tools, and coordinate with each other.

When to Use

Use this skill when:

Building new Python agents
Adding agentic capabilities to existing code
Integrating MCP tools with agents
Implementing multi-agent orchestration
Debugging agent behavior

Quick Start

Agentic Function (simplest)

from agentica import agentic

@agentic()
async def add(a: int, b: int) -> int:
    """Returns the sum of a and b"""
    ...

result = await add(1, 2)  # Agent computes: 3

Spawned Agent (more control)

from agentica import spawn

agent = await spawn(premise="You are a truth-teller.")
result: bool = await agent.call(bool, "The Earth is flat")
# Returns: False

Core Patterns

Return Types

# String (default)
result = await agent.call("What is 2+2?")

# Typed output
result: int = await agent.call(int, "What is 2+2?")
result: dict[str, int] = await agent.call(dict[str, int], "Count items")

# Side-effects only
await agent.call(None, "Send message to John")