#401 · Primary category: AI Coding Assistants
vibecode-pro-max-kit
Spec-driven coding harness with self-improving context memory, 15 agents, and 33 skills for autopilot workflows in Claude Code and Codex.
Project last updated:06/21/26
GitHub Stars
1.1K
Forks
231
Contributors
1
License
MIT
Why we included this project
Most AI coding agents lose the plot between sessions, so the harness's real value is keeping a long task on rails. It wraps Claude Code, Codex, Cursor, or any other agent in a seven-phase process that runs from research through spec to execution, and writes progress notes to disk at each step so a memory reset mid-task doesn't lose the thread. The /goal token keeps the agent running until a feature is done, self-healing loops re-check their own fixes up to ten times, and the setup scan builds project context and keeps docs current as you ship. Product owners can write requirements as plain-language user stories before any code exists, while engineers mostly get less context rot and fewer re-derived conclusions on larger codebases.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
superpowers
An agentic skills framework & software development methodology that works.
ECC
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
skills
Skills for Real Engineers. Straight from my .agents directory.
opencode
The open source coding agent.
andrej-karpathy-skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.