context-driven-development

Manages project context artifacts like product.md, tech-stack.md, and workflow.md for consistent AI-assisted development.

2|Updated Jun 16, 2026
One-click install
npx skills add https://github.com/monang404/lunawave --skill context-driven-development-monang404
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: context-driven-development
Source: https://github.com/monang404/lunawave/tree/main/.agent/skills/context-driven-development
Command: npx skills add https://github.com/monang404/lunawave --skill context-driven-development-monang404

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI-assisted development sessions often produce inconsistent results because project context lives in scattered prompts and tribal knowledge. This Skill establishes persistent, structured context artifacts (product.md, tech-stack.md, workflow.md, tracks.md) that keep AI behavior and team decisions aligned across sessions. ## Core Features & Use Cases - Context Artifact Management: Defines the purpose, contents, and update triggers for product.md, product-guidelines.md, tech-stack.md, workflow.md, and tracks.md. - Greenfield and Brownfield Setup: Guides creation of context from scratch for new projects or extraction of implicit context from existing codebases. - Validation and Anti-Pattern Detection: Provides checklists to detect stale context, context sprawl, and over-specification before starting implementation work. - Use Case: A team onboarding a new developer or AI assistant to an existing project uses this Skill to generate and validate conductor/ artifacts, ensuring every session starts from the same documented product vision, tech stack, and workflow rules. ## Quick Start Ask the AI to set up Conductor context artifacts for this project and validate that product.md, tech-stack.md, and workflow.md are current before starting the next track.

Frequently Asked Questions about context-driven-development

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

FAQPage Schema
How do I set up context-driven development for an existing codebase?▼

Run the Conductor setup flow in brownfield mode, which analyzes existing code, configs, and documentation to pre-populate artifacts. Then review and refine the generated product.md, tech-stack.md, and workflow.md to reconcile discovered patterns with desired standards.

What files does the Conductor context structure require?▼

The conductor directory contains index.md, product.md, product-guidelines.md, tech-stack.md, workflow.md, tracks.md, setup_state.json, plus code_styleguides and per-track directories with spec.md, plan.md, and metadata.json.

When should I update tech-stack.md during development?▼

Update tech-stack.md whenever adding dependencies, upgrading major versions, changing infrastructure, or adopting new tools. Document the rationale, version constraints, and configuration requirements before introducing any new dependency.

How does context-driven development keep AI sessions consistent?▼

Each session starts by reading index.md, tracks.md, and the active track's plan.md, so the AI works from the same documented context every time. Ending a session requires updating progress and status, preserving continuity across interruptions.

What are common context management anti-patterns to avoid?▼

The documented anti-patterns are stale context, context sprawl across multiple locations, implicit context never written down, context hoarding by one person, and over-specification that becomes unmaintainable. Each has a defined mitigation such as updating artifacts per track and reviewing context in pull requests.