learnship

Provide a multi-platform agent harness with persistent memory and a structured phase loop.

59|5|Updated Mar 8, 2026

Search Tags:#workflow #learning #ai agent #development process #harness #coding assistant

Ready to run

npx skills add https://github.com/FavioVazquez/learnship --skill learnship

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: learnship
Source: https://github.com/FavioVazquez/learnship/tree/main
Command: npx skills add https://github.com/FavioVazquez/learnship --skill learnship

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill provides a robust agent harness for building reliable AI coding agents, offering persistent memory, structured processes, and built-in learning capabilities to enhance productivity and understanding.

Core Features & Use Cases

Persistent Memory: AGENTS.md ensures agents retain project context across sessions.
Structured Process: A phase loop (Discuss → Plan → Execute → Verify) guides development.
Built-in Learning: Checkpoints at phase transitions foster deeper understanding.
Use Case: Start a new project with /new-project, manage your workflow with /ls and /next, and ensure UI quality with the impeccable design system.

Quick Start

Use the learnship skill to start a new project by typing /new-project.

Frequently Asked Questions about learnship

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I maintain persistent memory for an AI coding agent across sessions?▼

Persistent memory for AI coding agents is maintained using an AGENTS.md file to store project context, ensuring agents retain information across different coding sessions without losing prior progress.

What is a structured phase loop for AI development workflows?▼

A structured phase loop for AI development workflows is a systematic process guiding agents through Discuss, Plan, Execute, and Verify stages to ensure reliable code generation and verification.

Can I use this agent harness with Gemini CLI and Claude Code?▼

Yes, this agent harness supports multiple platforms including Gemini CLI, Claude Code, Windsurf, OpenCode, and Codex CLI, allowing flexible integration across various AI coding environments.

How do I start a new project with an AI coding assistant?▼

To start a new project with an AI coding assistant, you can use the /new-project command to initialize the workflow, followed by /ls and /next to manage the development process efficiently.

How do built-in learning checkpoints improve an AI coding workflow?▼

Built-in learning checkpoints occur at phase transitions within the development loop, fostering deeper understanding of the project context and improving the overall AI coding workflow efficiency.

What is the best way to enforce design standards in AI-generated code?▼

To enforce design standards in AI-generated code, you can utilize the impeccable design system integrated within the agent harness, ensuring UI quality and consistent design compliance.