DocsUseful Pluginsre_act

re_act

ReAct (Reason + Act) pattern for intelligent agents

Download

What it does

The re_act plugin implements the ReAct pattern:

Plan (after_user_input)

Before taking any action, the agent plans what to do based on user input and available tools.

Reflect (after_tools)

After each tool execution, the agent reflects on results and decides next steps.

Quick Start

main.py

from connectonion import Agent
from connectonion.useful_plugins import re_act

def search(query: str) -> str:
    """Search the web for information."""
    return f"Results for '{query}': Python is a programming language..."

agent = Agent("assistant", tools=[search], plugins=[re_act])

agent.input("Search for Python and explain what it is")

Python REPL

Interactive

💭 I'll search for information about Python, then explain what it is based on the results.

[Tool: search("Python")]

💭 The search returned that Python is a programming language. I now have enough context to explain.

Python is a high-level, interpreted programming language known for its simplicity...

How it works

1. Planning Phase

After receiving user input, the plugin generates a brief plan:

main.py

# Internal: plan_task handler
@after_user_input
def plan_task(agent):
    user_prompt = agent.current_session.get('user_prompt', '')
    tool_names = agent.tools.names()

    plan = llm_do(
        f"User request: {user_prompt}\nAvailable tools: {tool_names}\nBrief plan:",
        model="co/gemini-2.5-flash"
    )

    agent.current_session['messages'].append({
        'role': 'assistant',
        'content': f"💭 {plan}"
    })

2. Reflection Phase

After tools execute, the plugin reflects on results:

main.py

# Uses the built-in reflect handler from useful_events_handlers
@after_tools
def reflect(agent):
    trace = agent.current_session['trace'][-1]

    if trace['type'] == 'tool_execution' and trace['status'] == 'success':
        reflection = llm_do(
            f"Result: {trace['result'][:200]}\nWhat did we learn?",
            model="co/gemini-2.5-flash"
        )

        agent.current_session['messages'].append({
            'role': 'assistant',
            'content': f"💭 {reflection}"
        })

Combined with Eval Plugin

For debugging and testing, combine with the eval plugin:

main.py

from connectonion import Agent
from connectonion.useful_plugins import re_act, eval

# ReAct provides planning, eval provides evaluation at completion
agent = Agent("assistant", tools=[search], plugins=[re_act, eval])

agent.input("Search for Python")
# Output includes:
# 💭 Plan: I'll search for Python information
# [Tool execution]
# 💭 Reflection: We learned Python is a programming language
# ✓ Evaluation: Task complete - found and understood Python info

Events Used

Event	Handler	Purpose
`after_user_input`	plan_task	Generate initial plan
`after_tools`	reflect	Reflect on tool results

Source

connectonion/useful_plugins/re_act.py

main.py

# The plugin is just a list of event handlers
re_act = [plan_task, reflect]

Useful Plugins

eval