Summary

This repository offers a practical playbook for effective human-AI collaboration across research, writing, reading, and coding. It targets individuals seeking to leverage AI as a productive colleague while maintaining human oversight and judgment, aiming to enhance efficiency without sacrificing understanding or critical thinking. The core benefit is enabling users to lead AI-driven workflows, ensuring AI amplifies capabilities rather than replacing essential human cognitive functions.

How It Works

The core philosophy treats AI as a collaborative partner rather than a mere tool, emphasizing that humans remain the primary driver of problem-solving and decision-making. It advocates for integrating AI into real-world workflows with low friction, prioritizing context (goals, materials, preferences, acceptance criteria) before AI execution. The approach focuses on solidifying effective processes into reusable "Skills" and workflows, encouraging regular review and subtraction to prevent inefficiency and the "efficiency illusion."

Quick Start & Requirements

This repository is primarily a collection of methodologies, prompts, and agent rules rather than a software package with a direct installation command. Users are directed to read the main article (docs/phd-ai-collab.md) for the full methodology and explore directories like prompts/ and skills/full/README.md for reusable workflows. No specific software prerequisites or hardware requirements are listed, implying it's designed for general AI accessibility.

Highlighted Details

• Human-Centric AI Collaboration: Prioritizes human judgment, problem formulation, and final decision-making over AI autonomy.
• Practical Workflows: Offers a catalog of independently maintained "Skills" and prompts for specific tasks like paper review, prompt polishing, and anti-AI writing.
• Context-First Approach: Stresses the importance of defining goals, materials, preferences, and acceptance criteria before engaging AI.
• Combating Efficiency Illusion: Aims to prevent users from creating a false sense of progress by outsourcing critical understanding and judgment to AI.

Maintenance & Community

Feedback and bug reports are encouraged via GitHub Discussions and Issues. The project appears to be a personal endeavor ("我在读博士期间持续更新") with a link to a "小红书转发版" suggesting a broader audience engagement.

Licensing & Compatibility

The repository's README does not explicitly state a software license. Users should exercise caution regarding the reuse of content and methodologies, especially in commercial or closed-source contexts, until a license is clarified.

Limitations & Caveats

This playbook focuses on methodology and principles rather than providing a specific, installable toolset. Its effectiveness relies on the user's ability to adapt the principles to their specific AI tools and workflows. The lack of an explicit license poses a potential adoption blocker for certain use cases.