Skip to content
This repository was archived by the owner on Aug 4, 2026. It is now read-only.

Repository files navigation

TRAE-Tips: My Advanced TRAE Workflow & Agent Engineering

🏆 Winner of the 2025 TRAE Global Best Practice Challenge

Author: Marco — Full-Stack Developer & AI Workflow Architect

Star the project


⚠️ IMPORTANT MAINTENANCE NOTICE

This repository is no longer maintained.

Due to a loss of trust in TRAE's management and their lack of transparency regarding platform decisions, I have decided to discontinue my collaboration with TRAE. As a result:

  • No further updates will be made to this repository
  • Issues will not be addressed - the repository is effectively archived
  • No new content will be added
  • 📂 This repository remains available as a reference for the community
  • 🔄 Feel free to fork and continue the work independently

I believe in transparency and trust between platform providers and their community. When those values are compromised, it's time to move on.


📚 Table of Contents & Navigation


📌 Description

This repository documents the full engineering workflow I used to build, automate and scale software development tasks through TRAE, combined with custom-built AI Agents, strict Rule Systems, and an optimized model selection strategy based on performance benchmarks.

The core of this work is detailed in our Main Whitepaper, which explains how TRAE becomes a real AI Engineering Team, how to orchestrate multi-agent execution, and how leveraging GLM-5 (z.ai) inside TRAE dramatically reduces cost while increasing output efficiency.


📖 Detailed Guides

  • 💡 Best Models for TRAE - A complete comparison of models (Gemini, GPT, Kimi) and which one to choose for frontend, backend, or refactoring.
  • SOLO + GLM-5: Best Combo - Comprehensive deep-dive on the most cost-effective TRAE setup using GLM-5 to save up to 100x on credits.
  • 📋 Rulesets Template - Ready-to-use rule templates and system prompts for deterministic agent behavior.
  • 💡 10 Best Tips - Quick, actionable tips for immediate implementation and workflow optimization.
  • 📉 Billing & Token Optimization - Detailed strategies for saving tokens, reducing costs, and maximizing your requests.

📡 Platform Status & News

  • 💰 Billing & Membership Updates - Everything you need to know about the transition from request-based to token-based usage.
  • 📈 New Billing Stats Report - Real-world technical data, model efficiency, and token burn analysis from intensive usage.
  • 🤖 Model Availability - Why certain models are restricted and what are the best alternatives.
  • 🐧 Linux Version Status - Latest info on the upcoming (and currently in internal testing) Linux release.

🔗 Explore More


📄 License

This project is licensed under the MIT License - see the LICENSE file for details.


📊 Project Status

Metric Status
Maintenance Discontinued
Issues ❌ Not monitored
Pull Requests ❌ Not accepted
Content Updates ❌ None planned
Repository 📂 Available as reference
Forks ✅ Allowed and encouraged

💬 Final Notes

Thank you to everyone who supported, starred, and contributed to this repository. The TRAE community is amazing, and the knowledge shared here will remain available for anyone who finds it valuable.

For those continuing with TRAE, I wish you the best in your journey. Always prioritize platforms that value transparency and trust their community.


Last Updated: 2026

This repository is now archived. Fork it to continue the work.

About

TRAE-Tips is a curated collection of practical tips, best practices, and advanced insights for working efficiently with the TRAE AI ecosystem. It includes usage guides, optimization strategies, common pitfalls, and real-world workflows to help developers and power users get the most out of TRAE tools.

Topics

Resources

Stars

52 stars

Watchers

1 watching

Forks

Releases

Contributors

Languages