bmad-agent-builder

Build, edit, and analyze AI agent skills through conversational discovery.

Updated Apr 2, 2026
One-click install
npx skills add https://github.com/amitmishra-byldd/Bmad-Note-Maker --skill bmad-agent-builder-amitmishra-byldd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-agent-builder
Source: https://github.com/amitmishra-byldd/Bmad-Note-Maker/tree/main/_bmad/bmb/bmad-agent-builder
Command: npx skills add https://github.com/amitmishra-byldd/Bmad-Note-Maker --skill bmad-agent-builder-amitmishra-byldd

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml>=6.0, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill guides creators through conversational discovery to design, build, edit, and quality-check AI agents that are outcome-driven rather than mechanically over-specified. It reduces wasted tokens and brittle behavior by enforcing persona-led prompts, selective scripting for deterministic work, and clear activation/headless patterns.

Core Features & Use Cases

  • Conversational Discovery & Build Process: Six-phase guided flow to capture intent, strategy, requirements, and produce a complete agent structure (persona, capabilities, optional memory, assets, and scripts).
  • Quality Analysis & Reporting: Deterministic lint scripts plus LLM scanners synthesize a capability dashboard, agent portrait, and ranked opportunities for improvement.
  • Script & Execution Guidance: Identifies script opportunities, PEP 723 patterns, headless execution contracts, and outputs ready-to-run artifacts in the configured skills folder.
  • Use Cases: Create a new agent from intent, convert/edit an existing agent to be outcome-driven, run automated quality scans, or enable headless builds in CI.

Quick Start

Invoke the builder and say: Create an agent that acts as a technical reviewer for pull requests and save it to the project's skills folder.

Frequently Asked Questions about bmad-agent-builder

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

FAQPage Schema
How do I build an AI agent from scratch using conversational discovery?

To build an AI agent from scratch, conversational discovery guides you through a six-phase flow to capture intent and produce a complete outcome-driven persona, capabilities, optional memory, and assets.

What is the best way to convert an existing AI agent to be outcome-driven?

Converting an existing AI agent to be outcome-driven involves editing its persona and capabilities to enforce persona-led prompts and selective deterministic scripting, reducing wasted tokens and brittle behavior.

Can I run automated quality analysis on AI agent personas and capabilities?

Yes, you can run automated quality analysis on AI agent personas and capabilities using deterministic lint scripts and LLM scanners that synthesize a capability dashboard and ranked improvement opportunities.

Does the agent builder support headless execution and PEP 723 metadata for scripts?

The agent builder supports headless execution by identifying script opportunities and PEP 723 patterns, outputting ready-to-run deterministic artifacts with proper execution contracts.

Do I need pyyaml to generate AI agent skills and SKILL.md frontmatter routing?

You need pyyaml (version 6.0 or higher) to support the underlying scripts and metadata processing required to generate AI agent skills and satisfy SKILL.md frontmatter routing requirements.