Launch an autonomous AI agent with sandboxed execution in 2 lines of code

OnPrem.LLM's AgentExecutor enables launching autonomous AI agents in 2 lines of code, supporting both cloud models (OpenAI, Anthropic, Gemini) and local models (Ollama, vLLM, llama.cpp) via LiteLLM. The agent comes with 9 built-in tools including file I/O, shell execution, and web search, all operating within a sandboxed working directory. Users can customize tool access, disable shell for safety, or add custom tools, with the underlying implementation powered by PatchPal.