py-learn

Creates a structured issue.md file in the project root, designed to be implemented by a junior programmer or a cheaper AI model. Use when the user asks to "create an issue", "write up a task", "create a ticket", or "document what needs to be done". The agent enriches the user's description with codebase context to produce a clear, self-contained, and unambiguous issue.

Execute the full ship flow: conventional commit → push → Pull Request. Use when user asks to commit changes, ship a feature, create a git commit, or mentions "/commit" or "/ship". Supports: (1) Auto-detecting type and scope from changes, (2) Generating conventional commit messages from diff, (3) Interactive commit with optional type/scope/description overrides, (4) Intelligent file staging for logical grouping, (5) Auto-push to remote branch, (6) PR creation via GitHub CLI (gh)

Implements the task described in issue.md by reading it thoroughly, proposing a plan for user approval, creating a new branch, then writing the code. Use when the user asks to "implement the issue", "work on the issue", "start the task in issue.md", or "implement what's in issue.md". Always stops before committing — the user reviews the code first.

Generates a production-ready Dockerfile for a Next.js project using pnpm, standalone output mode, and self-hosted VPS deployment. Use when the user asks to "generate a Dockerfile", "dockerize the project", "create a Dockerfile for Next.js", or "set up Docker for production". Also triggers when the user wants to containerize or deploy the Next.js app to a VPS.