project-builder

Guide structured reflection on Carta dogfooding sessions to produce todo tasks.

1|Updated Jan 19, 2026
One-click install
npx skills add https://github.com/saxonthune/carta --skill project-builder-saxonthune
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: project-builder
Source: https://github.com/saxonthune/carta/tree/main/.claude/skills/project-builder
Command: npx skills add https://github.com/saxonthune/carta --skill project-builder-saxonthune

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Helps users reflect on building a real (non-Carta) project with Carta so they can identify what’s missing, reduce friction, and translate that experience into actionable improvements.

Core Features & Use Cases

  • Dogfooding reflection workflow: guides the user to describe what they’re modeling, what went wrong or felt awkward, and what would make the experience better.
  • Carta state investigation via MCP/docs: prompts targeted checks of Carta’s current capabilities using the workspace and relevant documentation.
  • Deterministic todo-task output: writes improvement tasks into todo-tasks/ only, capturing origin, design intent, and verifiability.

Quick Start

Invite the skill while you’re building a non-Carta project in Carta and ask it what would make that modeling experience better for your specific workflow.

Frequently Asked Questions about project-builder

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

FAQPage Schema
How do I turn dogfooding friction into actionable product improvements?

Dogfooding friction converts into actionable improvements through structured reflection that identifies workflow gaps during external project modeling, then generates targeted investigation tasks capturing design intent and verifiability for product enhancement.

Can I generate todo tasks from UX research sessions without editing source code?

Todo tasks generate from UX research without editing source code by using conversation plus MCP and documentation context to produce implementation-ready tasks, writing exclusively to the todo-tasks directory while never reading external project source implicitly.

What's the best way to identify gaps when modeling external CLI workflows?

Identify gaps during CLI workflow modeling by applying structured reflection and performing targeted checks of current capabilities using workspace and documentation context, pinpointing missing features and translating friction into verifiable next-step enhancements.

Does this MCP integration approach work for documenting architecture friction?

MCP integration supports documenting architecture friction by prompting targeted checks of current capabilities using workspace and documentation context, guiding structured reflection on what went wrong and capturing verifiable improvement tasks in todo-tasks.

Are there limitations on what files the dogfooding reflection workflow can modify?

Limitations restrict file modifications to the todo-tasks directory only, enforcing safety constraints by never editing source code and never reading external project source implicitly, ensuring all generated improvement tasks remain isolated and verifiable.