bmad-build

Implements features and bug fixes through a spec-driven plan, implement, and review workflow.

Updated Mar 1, 2026
One-click install
npx skills add https://github.com/Hexalith/Hexalith.Parties --skill bmad-build-hexalith
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-build
Source: https://github.com/Hexalith/Hexalith.Parties/tree/main/.agents/skills/bmad-build
Command: npx skills add https://github.com/Hexalith/Hexalith.Parties --skill bmad-build-hexalith

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Turning a feature request or bug report into verified, reviewed code requires disciplined planning, scoped specifications, and independent review — steps that are easy to skip and hard to keep consistent. This Skill enforces that discipline as a structured workflow. ## Core Features & Use Cases - Spec-driven implementation: Clarifies intent, investigates the codebase, and writes a token-bounded spec with code maps, acceptance criteria, and open questions before any code changes. - Multi-layer automated review: Runs blind-hunter, edge-case-hunter, and verification-gap reviewer subagents against the staged diff, then triages every finding into patch, spec-fix, or defer routes. - Epic and sprint integration: Compiles epic context from planning artifacts and syncs story status with sprint-status.yaml for BMAD-managed projects. - Use Case: Delegate a story like "implement story 3-2 digest delivery" and receive a planned spec, implemented diff, reviewed findings, and a conventional commit. ## Quick Start Ask the agent to implement a feature, story, or bug fix and it will plan, build, review, and commit the change through the bmad-build workflow.

Frequently Asked Questions about bmad-build

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

FAQPage Schema
How do I implement a feature or bug fix with an AI coding workflow?

Delegate the feature, story, or bug fix as a request and the workflow clarifies intent, investigates the codebase, writes a spec, implements it via a subagent, and runs layered reviews before committing. A bare story or issue link is enough to start.

How does automated code review work in this workflow?

Three review layers run against the staged unified diff: a blind hunter, an edge-case path tracer, and a verification-gap reviewer. Every finding is verified against the code, assigned a verdict, and routed to patch, spec amendment, or deferral.

When should I not use bmad-build for a change?

Skip it for obvious low-risk mechanical maintenance such as small ignore-file edits, typo-only fixes, formatting-only changes, or configuration hygiene. It also does not handle user-directed interactive edits or version-control operations that only record existing work.

Does bmad-build support epic and sprint tracking?

Yes. It compiles epic context files from planning artifacts, resolves story keys, and syncs development status in sprint-status.yaml as stories move to in-progress and review states.

What happens when a review finding cannot be fixed by a small patch?

Findings caused by incomplete intent loop back to the human, while spec-level defects trigger a spec amendment and code re-derivation with a change log entry. Loopbacks are capped at five iterations before escalating to the human.