mini_agent
Active·★ 37·Updated 2026-08-31
Mini Agent is a minimal AI agent built from scratch using Python, OpenAI SDK, and a simple while loop, without relying on complex frameworks. It takes a task, dynamically calls tools, observes results, and iterates until a final answer is achieved, supporting local, remote, and mixed execution modes.
#AI Agent#Python#LLM Orchestration#Tool Use#Reliability#Ollama#Visual Testing#CI/CD Automation
01
Features
01Flexible LLM Execution Modes (Local, Remote, Mixed)
02Robust Reliability Layer (Retry, Circuit Breaker, Validation, Tracing)
03Extensible Tool Calling Mechanism with JSON Schema
04Visual Testing Agent for Mobile UI Automation
05Integrated Deployment Pipeline for CI/CD Automation
02
Compatibility
Python
Runtime
Verified via docs
Ollama
Local LLM
Verified via docs
OpenAI
Cloud LLM
Verified via docs
Anthropic
Cloud LLM
Verified via docs
Google Gemini
Cloud LLM
Verified via docs
iOS Simulator
Visual Testing Target
Verified via docs
03
Quick start
1
$ git clone https://github.com/sergenes/mini_agent
2
$ cd mini_agent
3
$ python -m venv .venv
4
$ source .venv/bin/activate # Windows: .venv\Scripts\activate
5
$ pip install -r requirements.txt
6
$ cp .env.example .env
04
Use cases
↳Automating multi-step tasks requiring dynamic tool use and reasoning (e.g., getting date/weather, calculating, file operations).
↳Developing and testing AI agents with a focus on core mechanics and reliability, without framework overhead.
↳Performing visual flow testing for iOS Simulator and Android applications, judging screens by visual intent.
↳Integrating AI-driven automation into CI/CD pipelines for mobile and web application deployment.
↳Experimenting with different LLM providers (OpenAI, Anthropic, Gemini, Ollama) and fallback strategies for agent tasks.
05
Alternatives
gpt2agent★ 52
Your codex login → a full ChatGPT Plus/Pro account (every model, deep research, image gen, code exec) inside Claude Code, Codex & any MCP client. One-line install.
Microsoft AutoGen★ 61.5k
A framework that enables the development of LLM applications using multiple agents that can converse with each other to solve tasks.
ptah-extension★ 29
Provider-agnostic AI orchestration for VS Code. Intelligent workspace analysis, Code Execution MCP server, and project-adaptive multi-agent workflows — all natively integrated into VS Code.
PraisonAI★ 9.3k
PraisonAI is a production-ready Multi AI Agents framework, designed to create AI Agents to automate and solve problems ranging from simple tasks to complex challenges. It provides a low-code solution to streamline the building and management of multi-agent LLM systems, emphasising simplicity, customisation, and effective human-agent collaboration.
intentkit★ 6.5k
An open and fair framework for everyone to build AI agents equipped with powerful skills. Launch your agent, improve the world, your wallet, or both!
Comments
Log in to leave a comment
No comments yet. Be the first!