Show HN: Ergo – Minimal, fast, persistent task backlog in your repo (github.com)

Ergo is a new command-line interface (CLI) tool designed to enhance task planning for AI development environments, specifically for Claude Code and Codex. This tool allows users to maintain a clear, persistent backlog of tasks and epics within their code repositories using a compact, git-friendly JSONL format. By separating the goals of what to build from the implementation processes, Ergo reduces the chaos often associated with AI agent planning, which can generate unwieldy markdown files that blur essential distinctions. This tool is particularly significant for the AI/ML community as it enables concurrent task management among multiple agents, improving efficiency in project execution. Ergo boasts a 5-15x speed advantage over previous tools, like beads, especially for larger projects, while ensuring safe, serialized writes through a simple file-locking mechanism. Its append-only JSONL format offers full audit trails, making it easy to track changes, recover from errors, and maintain the integrity of tasks. With built-in support for prioritization, task dependencies, and automated validation, Ergo aims to streamline the planning and implementation phases in AI-driven software projects, making it a powerful asset for developers.