Endy – orchestrate multiple coding-agent CLIs to cut LLM spend (github.com)

Endy has launched a new local control plane designed for coding agents, enabling developers to orchestrate various agent CLIs like Codex, OpenCode, CommandCode, and Hermes within a single tmux session. This tool facilitates streamlined project management by allowing users to manage multiple coding tasks efficiently from a unified command interface. Each project is organized within its own tmux session, providing a global overview of all projects, durable task logs, and simplified task delegation through commands like `endy spawn`. This development is significant for the AI/ML community as it addresses the growing need to optimize costs associated with using large language models (LLMs) by providing a more effective way to manage concurrent coding tasks. By logging task outputs and statuses, developers can easily track progress and improve productivity without excessive LLM spending. The endy tool’s requirements are straightforward — it runs on macOS or Linux, needs Python 3 and tmux, and can work with just one of the supported agent CLIs installed, making it accessible for a broad audience of developers eager to enhance their workflow with advanced AI coding capabilities.