ИИ-агенты
Talorys
Talorys — your personal AI agent in your own Cloudflare account: chat, memory, tasks, notes and scheduled routines, deployed with one command
★ 1,1 тыс. звёздMIT
Смотреть руководство →FrontierAgent — проект с открытым исходным кодом из категории «ИИ-агенты» от ApodexAI (ApodexAI/FrontierAgent на GitHub). У него 5 165 звёзд на GitHub и 278 форков, он распространяется по лицензии Apache-2.0 и последний раз обновлялся на GitHub 8 октября 2026.
An open agent framework with a native terminal TUI, ReAct and Agent Team modes — one command to run.
Define the task you want executed — research, documents, code — and which workflow mode fits: solo ReAct or an Agent Team.
Clone the repo and sync the lightweight terminal runtime with uv for Python 3.12.
Copy .env.example to .env and add your OpenAI-compatible endpoint and key (and a search provider key if you want web search).
Start FrontierAgent in your project, watch the plan board and tool activity, and steer with follow-ups as it works.
A full terminal interface with a sidebar plan/task board, live tool activity, deliverables and a session-scoped diff.
Choose a single-agent ReAct loop or an Agent Team workflow, depending on the task's size and parallelism.
Works with any OpenAI-compatible model endpoint configured in .env; hosted endpoint install scripts are provided for macOS and Linux.
Built-in web_search supports Serper-compatible endpoints or the free, keyless Parallel Search MCP, with web_fetch for reading pages.
While a workflow is busy, follow-up messages queue for the next safe turn boundary instead of interrupting the run.
Docker deployment and native or containerized SGLang on Linux NVIDIA hosts are documented for self-hosted model serving.
Clone FrontierAgent and enter the directory.
git clone https://github.com/ApodexAI/FrontierAgent.git
cd FrontierAgentInstall the terminal runtime with uv and create your .env with an OpenAI-compatible endpoint.
uv sync --python 3.12 --extra dev
cp .env.example .env
# Set OPENAI_API_KEY and OPENAI_BASE_URL in .envStart FrontierAgent in your project and give it a task in the TUI.
uv run frontier-agent # see README for the exact commandFrontierAgent is maintained by ApodexAI, open-sourced alongside its agent product.
Yes — the framework is Apache-2.0 licensed. You pay for the model endpoint you connect.
You need an OpenAI-compatible model endpoint — a hosted provider key or your own self-hosted endpoint (for example SGLang) both work. Web search optionally uses a Serper key, or the free keyless Parallel option.
Those are coding-focused terminal agents; FrontierAgent is a general agent framework with its own TUI, ReAct and multi-agent Team modes, and first-class document/scientific workflows.
Clone the repo, uv sync, copy .env.example to .env with your endpoint, and launch the TUI.
Talorys — your personal AI agent in your own Cloudflare account: chat, memory, tasks, notes and scheduled routines, deployed with one command
A portable AI agent runtime for the JVM — write one @Agent, run it on any Java AI stack.
Multi-agent LLM framework that mirrors a trading firm: analysts, researchers, trader and risk team
ZCode — Z.ai's AI coding workspace: desktop app, browser interface and terminal agent in one open-source harness
A framework for building agentic apps
One portable, local-first memory layer for every AI agent - Markdown-native, user-owned and self-evolving.
Другие проекты из категории «ИИ-агенты» с наибольшим числом звёзд в каталоге LibreAI — реалистичный список альтернатив, если FrontierAgent вам не подходит.
A personal AI that really does things — on any OS, with your data staying yours.
An agentic skills framework & software development methodology that works.
Agent harness optimization system with skills, memory and security for Claude Code, Codex and more
The agent that grows with you — a Python framework for adaptive AI agents.
DeepSeek's plugin-based agent harness where everything is a plugin.
Соседние проекты из категории «ИИ-агенты» в каталоге, ближайшие по числу звёзд к FrontierAgent — продолжайте изучать полку по обе стороны от него.
Connect your Zotero research library to Claude and other AI assistants through MCP.
Talk to your computer — an open-source voice interface.
The harness of harnesses: a persistent, self-evolving multi-agent ecosystem built for RSI.
Exa's MCP server for web search and crawling — give your AI agent a live, high-quality web index.
An open-source platform for AI-native application development — assistants, tools, RAG and models in one API.
The open-source AgentOS, formerly known as AutoGen.
Бесплатно и открыто — забирайте исходный код, ставьте звезду репозиторию и начинайте создавать.