Search tldr

GitHub - snarktank/ai-dev-tasks: A simple task management system for managing AI dev agents · GitHub

AI Dev Tasks is a lightweight prompt-based workflow for directing AI coding agents through requirements, granular planning, and reviewable implementation steps. It replaces one-shot feature prompts with a PRD-to-task-list process that keeps humans in control.

Share

GitHub - snarktank/ai-dev-tasks: A simple task management system for managing AI dev agents · GitHub

Author: snarktank | Published: Unknown | Generated: 2025-06-12 | Domain: github.com
Tags: ‘#ai-agents’ ‘#ai-coding’ ‘#task-management’ ‘#prd’ ‘#developer-workflow’


TLDR

AI Dev Tasks provides two reusable Markdown prompts—one to produce a Product Requirements Document (PRD), another to turn it into a detailed task list—for managing AI-assisted feature development. Rather than issuing a monolithic coding request, developers define scope, generate granular tasks, and ask the agent to complete and mark off work one sub-task at a time. The workflow emphasizes human review, explicit context, and iterative correction to improve reliability on larger features.

Key Takeaways

  • Three-stage workflow: Define a feature in a PRD, convert the PRD into an actionable task list, then have the AI implement individual sub-tasks with review checkpoints.
  • Two core prompt files: create-prd.md guides requirement creation, while generate-tasks.md decomposes a PRD into implementation work.
  • Human-controlled iteration: Developers review each small change before continuing, reducing the risk that an agent loses context or produces overly complex, incorrect code.
  • Tool-agnostic design: The Markdown-based prompts are intended for AI coding assistants including Amp, Claude Code, and Windsurf.
  • Repository traction: At the captured time, the public repository reports 7.8k stars, 1.7k forks, 218 watchers, 65 commits, and 12 contributors.

Images & Media

Keep reading