Summary

dcramer/dex provides a structured task-tracking system designed for AI agents, offering persistent memory across complex, multi-session workflows. It targets developers building or integrating AI agents, enabling robust coordination and state management beyond simple to-do lists by capturing full task context.

How It Works

The system defines tasks with distinct Description, Context, and Result fields, mirroring issue tracking workflows. This structured approach facilitates agent coordination by allowing agents to break down work, track progress, and record outcomes. Data is stored in a Git-friendly JSONL format (one task per line), promoting collaboration, versioning, and conflict-free merging.

Quick Start & Requirements

Install globally via npm install -g @zeeg/dex or integrate with AI agents using npx skills add dcramer/dex. Supports agents like Claude Code, OpenCode, Codex, and Cursor. Development requires Node.js and pnpm; local testing involves running the AI agent with the plugin directory specified (claude --plugin-dir plugins/dex). No specific hardware or OS prerequisites are detailed.

Highlighted Details

• Captures full task context (Description, Context, Result) for detailed state management.
• Utilizes Git-friendly JSONL storage for version control, collaboration, and conflict resolution.
• Designed specifically for agent coordination and persistent memory across sessions.
• Integrates with popular AI agent platforms.

Maintenance & Community

Release automation is handled via GitHub Actions upon pushing version tags. The provided README does not detail specific community channels (e.g., Discord, Slack), active contributors, sponsorships, or a public roadmap.

Licensing & Compatibility

The repository's license is not explicitly stated in the provided README text. This omission requires clarification for commercial use or integration into closed-source projects.

Limitations & Caveats

The project appears to be under active development, with setup instructions focused on installation and local testing. No specific limitations, known bugs, or alpha status are mentioned. The absence of a clear license is a significant adoption blocker.