Agent Development
Build AI agents using the Spoon AI ReAct framework.
Agent Hierarchy
code
SpoonReactSkill → Skills + x402 Payments
↓
SpoonReactMCP → MCP Tool Integration
↓
SpoonReactAI → Base ReAct + Tool Calling
↓
ToolCallAgent → Tool Execution
↓
BaseAgent → Pydantic Foundation
Quick Start
python
from spoon_ai.agents import SpoonReactMCP
from spoon_ai.chat import ChatBot
from spoon_ai.tools import ToolManager
from spoon_ai.tools.mcp_tool import MCPTool
# Create agent with MCP tool
agent = SpoonReactMCP(
name="my_agent",
llm=ChatBot(model_name="gpt-4o"),
tools=ToolManager([MCPTool(...)]),
max_steps=15
)
await agent.initialize()
result = await agent.run("Your query here")
Scripts
| Script | Purpose |
|---|---|
| basic_agent.py | Simple SpoonReactAI agent |
| mcp_agent.py | MCP-enabled agent |
| custom_tool.py | Custom BaseTool implementation |
| concurrent_agents.py | Multi-agent parallel execution |
References
| Reference | Content |
|---|---|
| prompts.md | System prompt templates |
| mcp_config.md | MCP configuration options |
Configuration
| Parameter | Type | Default | Description |
|---|---|---|---|
name | str | "spoon_react" | Agent identifier |
max_steps | int | 10 | Max ReAct iterations |
tool_choice | str | "required" | "auto", "required", "none" |
system_prompt | str | None | Custom system prompt |
Best Practices
- •Pre-load MCP tool parameters before use
- •Use concurrent execution for independent tools
- •Implement retry with exponential backoff
- •Keep system prompts focused and specific
- •Never hardcode API keys