ai-development-loop

Automate ticket-driven development with atomic updates and quality gates.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/levonk/dotfiles --skill ai-development-loop
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ai-development-loop
Source: https://github.com/levonk/dotfiles/tree/main/home/current/.chezmoitemplates/config/ai/skills/software-dev/ai-development-loop
Command: npx skills add https://github.com/levonk/dotfiles --skill ai-development-loop

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires tkr, just, direnv, devbox, mise, nix, poetry, pipenv, uv, pyenv, asdf, docker, rg, npx, pnpm, yarn, npm, cargo, playwright, and includes scripts (resource) components.

What problem does it solve?

The AI Development Loop provides a repeatable, quality-focused workflow for agents that manage ticketr-based projects, ensuring consistent planning, execution, and reflection.

Core Features & Use Cases

  • Atomic ticket updates on start and completion to maintain accurate progress
  • 9-step development loop: foundation check, ticket selection, start, high quality, strategy, implementation, verification, ticket audit, completion, commit
  • Integration with the ticketr CLI and auto-generated templates for boilerplate improvements
  • Use cases include starting new features, conducting infrastructure updates, and documenting learnings from reflections

Quick Start

  • Get ready to work with the next ticket using tkr ready, then begin the cycle with the dev-loop workflow (foundation, start, implement, verify, reflect).

Frequently Asked Questions about ai-development-loop

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

FAQPage Schema
How do I automate ticket-driven development workflows for AI projects?

Ticket-driven development is automated through a 9-step loop covering foundation checks, ticket selection, implementation, verification, and reflection to enforce consistent AI project quality. It integrates with the ticketr CLI to manage atomic ticket updates.

What is the AI development loop process for managing tickets?

The AI development loop is a 9-step cycle: foundation check, ticket selection, start, high quality, strategy, implementation, verification, ticket audit, completion, and commit. It enforces quality gates and tracks reflection opportunities for continuous project improvement.

How do I enforce quality gates in an automated software development cycle?

Quality gates are enforced during the development cycle by requiring verification and ticket audits before completion. The workflow integrates environment-aware execution and atomic ticket updates to maintain standards across foundation, implementation, and reflection phases.

Do I need the ticketr CLI to use a ticket-driven development workflow?

Yes, the ticketr CLI is required as the workflow integrates directly with it for ticket selection and auto-generated templates. The loop uses tkr ready to prepare the next ticket and manages atomic ticket updates on start and completion.

Can I standardize AI agent development cycles across different environment managers?

AI agent development cycles are standardized through environment-aware execution supporting multiple managers. The workflow adapts to your configured environment using dependencies like docker, direnv, devbox, mise, nix, poetry, and uv for consistent project execution.

What's the best way to track reflections and learnings from automated development loops?

Reflections and learnings are tracked through a formal reflection and opportunity-tracking process built into the development loop. After verification and completion, the workflow documents learnings to improve future ticket-driven cycles.