AuC
Agents-ufy-Core
AuC (Agents-ufy-Core) is a Python asyncio-based AI agent framework that provides a pluggable reasoning loop with ReAct capabilities, multi-LLM support (OpenAI, Anthropic, DeepSeek), and tiered tool permissions (L1/L2/L3). It incorporates Claude Code-inspired engineering practices including context slicing, project rules via `.aurules` files, and multi-channel approval workflows for high-risk operations. The framework ships with both terminal REPL and web UI modes, supports checkpoint rollback, evolution memory for self-improvement, and integrates with Telegram and QQ for remote authorization workflows.
Key Features
Use Cases
- 01Building autonomous coding agents with safe file system and shell access through tiered permissions
- 02Running AI agents in sandboxed environments with Docker isolation and resource constraints for production workloads
- 03Implementing remote approval workflows for high-risk agent operations via Telegram or QQ messaging platforms
- 04Creating self-improving agents that learn from execution history through evolution memory and metric tracking
- 05Developing multi-model AI systems with intelligent routing across OpenAI, Anthropic, and DeepSeek based on cost, quality, or latency
- 06Prototyping agentic applications with code indexing, multi-turn reviews, and task replay capabilities
Related Agents
View morehermes-agent
The agent that grows with you
agency-agents
A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.
openinterpreter
A coding agent for open models like Kimi K3
cline
Autonomous coding agent as an SDK, IDE extension, or CLI assistant.
AuC — FAQ
What is AuC and what does it do?+
AuC (Agents-ufy-Core) is a Python framework for building autonomous AI agents with ReAct reasoning, multi-LLM support, tiered tool permissions, and Claude Code-inspired safety practices. It provides both terminal and web interfaces for agent interaction and includes checkpoint rollback, evolution memory, and remote authorization workflows.
How do I install AuC?+
Install from PyPI using `pip install "ufy-auc[chat]"` for terminal mode or `pip install "ufy-auc[web]"` for web UI. The package name is `ufy-auc` on PyPI (import as `auc`). You can also install from source by cloning the GitHub repo and running `pip install -e ".[chat]"`.
What API keys and prerequisites does AuC require?+
AuC requires Python 3.11+ and at least one LLM provider API key: `OPENAI_API_KEY`, `ANTHROPIC_API_KEY`, or `DEEPSEEK_API_KEY`. Configure providers using `auc config init` which creates `~/.Au/AuC/settings.json`. Optional dependencies include Docker for container isolation and Telegram/QQ bot tokens for remote authorization.
Which AI clients or environments work with AuC?+
AuC is a standalone agent framework, not a plugin for other clients. It runs independently via its own `auc chat` terminal interface or `auc web` browser UI. It can integrate with external systems through its MCP client support and approval ports.
Is AuC free and open source?+
Yes, AuC is released under the MIT license and is completely free to use. The source code is available on GitHub at https://github.com/ufy2024/AuC.
How does the L1/L2/L3 permission system work?+
L1 tools are read-only (safe by default), L2 tools can write within a sandbox, and L3 tools require secondary authorization through Telegram, QQ, web UI, or console. Dangerous shell commands automatically escalate to L3, and approval modes can be configured to always ask, ask only for dangerous operations, or auto-approve.
How do I install AuC?+
Open the source repository on GitHub and follow its README. AuC is a agent — MCP Agents Market links you directly to the official repo.
Is AuC free?+
AuC is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.