archon-dev

Route development tasks into deterministic Archon workflows from research through PR creation.

Updated May 21, 2026
One-click install
npx skills add https://github.com/sunzcdev/archon-dev-team --skill archon-dev-sunzcdev
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: archon-dev
Source: https://github.com/sunzcdev/archon-dev-team/tree/main/.claude/skills/archon-dev
Command: npx skills add https://github.com/sunzcdev/archon-dev-team --skill archon-dev-sunzcdev

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes assets (resource) components.

What problem does it solve?

The archon-dev skill solves the problem of inconsistent, non-repeatable AI-driven software development by routing every development task (research through PR creation) into a deterministic, evidence-based workflow with clear phases and guardrails.

Core Features & Use Cases

  • Intent-based routing to specialized cookbooks: Automatically selects the right workflow for research, investigate, PRD writing, planning, implementation, review, debugging, committing, PR creation, or issue reporting.
  • Evidence-first development rules: Enforces “file:line” evidence for codebase claims and prohibits speculation when uncertain.
  • Artifact-driven workflow structure: Writes standardized outputs into .claude/archon/ (plans, reports, issues, reviews, debug, research) to keep work traceable and composable.

Quick Start

Tell the skill what you are trying to do, for example: send “plan migrate caching layer” to create an implementation plan for the requested change.

Frequently Asked Questions about archon-dev

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

FAQPage Schema
How do I enforce evidence-based code review in an AI development workflow?

Evidence-based code review enforces file:line citations for codebase claims, routing AI coding tasks into deterministic workflows that prohibit speculation and require traceable artifact generation.

What is the best way to plan an implementation before creating a GitHub pull request?

Planning an implementation before a GitHub pull request involves routing the task into a deterministic workflow that generates standardized requirement and planning artifacts within the .claude/archon/ directory.

How do I route coding agent tasks through deterministic development phases from research to PR creation?

Routing coding agent tasks through deterministic development phases uses intent-based selection to match tasks like research, planning, or debugging to specialized cookbook workflows ending in GitHub PR creation.

Can I use artifact management to make AI codebase discovery traceable and composable?

Artifact management makes AI codebase discovery traceable by writing standardized outputs for plans, reports, and research into the .claude/archon/ directory, ensuring every workflow phase is composable.

Why does my AI coding agent speculate about codebase changes instead of applying evidence-based debugging?

AI coding agents speculate without evidence-based debugging guardrails, which enforce file:line evidence for codebase claims and route investigations into deterministic workflows that prohibit uncertainty.

Do I need a specific directory structure to manage development workflow artifacts for coding agents?

Managing development workflow artifacts requires the .claude/archon/ directory structure to store standardized outputs like plans, reviews, and debug reports, keeping coding agent workflows traceable.