tanweai/pua
- Source
- GitHub
- First trending
- Category
- Skills
- GitHub stars
- 19,652
- Main language
- Python
This page introduces an external open-source repository. It is not an HDATF product.

What it does
A skill plugin for coding agents such as Claude Code and Codex. It combines workplace pressure language modeled on a performance improvement plan, a debugging method and proactivity rules to push the agent to exhaust options before giving up.
How it helps ATF
Its rules against giving up early could be worth comparing with how Harness checks results with evidence. The project itself notes that a productivity gain has not been established.
License
No license No license file found. Without a license, reuse of the code is not permitted by default.
More in this category
- garrytan/gstack
gstack packages Garry Tan's exact Claude Code setup: 23 opinionated tools that act as CEO, designer, engineering manager, release manager, doc engineer and QA. - pbakaus/impeccable
A skill giving AI coding agents frontend design guidance. An init step records product facts in PRODUCT.md, commands such as polish, audit and critique share a design vocabulary, and deterministic detector rules flag typical AI-generated design patterns. - mattpocock/skills
A personal set of small, composable agent skills for everyday engineering work, easy to adapt and usable with any model. It can be installed as a managed Claude Code plugin or copied as editable files through skills.sh. - phuryn/pm-skills
A marketplace of product management skills, commands and plugins for Claude Code and Cowork, from discovery and strategy to PRD writing, launch and growth. Each skill walks the user through a PM framework step by step. - MiniMax-AI/skills
A beta collection of development skills for AI coding agents, giving structured guidance for frontend, fullstack, Android, iOS and shader work. The frontend skill can also generate media assets through the MiniMax API.
Only repositories in the ranked Trendshift lists are included, and the lists are used only to find candidates. We do not copy their ranks. Descriptions, licenses and star counts come from each GitHub repository. The notes are our own reading. We have not tested these projects, and a place on a trending list does not prove quality.