sheikhahmed2606-web/ECC — reverse-engineered prompt

I want a repo that feels like an operating system for AI coding agents, not just a pile of prompts. Build a cross tool setup that works with Claude Code, Codex, Cursor, OpenCode, Gemini, Zed, and similar agents, with reusable skills, rules, hooks, commands, memory that carries context across sessions, and a research first workflow so the agent checks docs and verifies things before making changes.

It should help with performance and token efficiency, support orchestrating bigger tasks, keep useful patterns from past sessions, and include security minded checks so unsafe or sloppy changes get caught early. I also want a simple desktop dashboard for managing sessions, configs, and workflows, plus install scripts and clear guides so someone can get started fast.

Please make it feel production ready, with examples, sane defaults, and compatibility for older command styles where needed. If anything is unclear, look up current docs online and use the README as the source of truth for the overall product.