bmad-agent-builder
OfficialBuild outcome-driven AI agents via guided discovery
System Documentation
What problem does it solve?
Helps creators and engineers rapidly design, convert, and quality-check AI agent skills by running a structured conversational discovery process and producing a ready-to-use skill bundle. It removes the manual toil of scaffolding persona, capabilities, memory sanctum templates, and headless execution paths so agents are consistent, outcome-focused, and IDE-ready.
Core Features & Use Cases
- Conversational Discovery & Build: Guided six-phase discovery that turns a user's vision into a stateless, memory, or autonomous agent with seeded sanctum templates.
- Edit, Analyze, Rebuild: Convert or improve existing agents through targeted analysis, pruning over-specification, and regenerating lean capability prompts.
- Headless & Reports: Support for --headless automation, init-sanctum scaffolding, and quality-analysis report generation for CI or scheduled runs.
- Use Case: Convert a verbose agent into an outcome-driven memory agent with seeded PERSONA/CREED/BOND files, capability prompts in references/, and an init script to create the sanctum.
Quick Start
Ask the builder to create or analyze an agent by saying a natural command like: create an agent that is a concise Python code reviewer with optional memory and headless report generation.
Dependency Matrix
Required Modules
Components
💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: bmad-agent-builder Download link: https://github.com/ImaginerLabs/skill-manager/archive/main.zip#bmad-agent-builder Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.