Skills bmad-code-org bmad-method BMad Method

bmad-build

Use it when:: You hand the agent a ticket, story, feature or bug fix and leave the edits to it.
It does:: Investigates the code, writes a plan file, implements it, has a reviewer subagent check the diff, fixes what's real and commits.
Two routes:: About 100 changed lines or fewer: built in the session; larger: you approve the plan, then a fresh subagent implements it.
You get:: A local commit, not pushed on its own, and a plan file with the intent and every review verdict; larger changes add a code map, tasks and acceptance criteria.
Unlike bmad-build-auto:: This one stops to ask you about gaps in the intent, a dirty working tree or a request with several goals.
Needs:: uv and BMad set up in the project: run the bmad skill's setup first; subagents for implementation and review.

Skill 12 of 24 in the BMad Method bundle, after bmad-agent-dev and before bmad-build-auto.

Only works inside its bundle: Its workflow is printed by a script in the _bmad/ folder that the bmad skill's setup creates, so it does nothing without that setup. Install BMad Method instead.

Needs uv and BMad set up in the project: it runs scripts from the _bmad/ folder that the bmad skill's setup creates.

TL;DR

BMad's build workflow takes a change from intent to a local commit. It starts from what you hand it: a ticket from BMad's ticket tree of initiative, epic and story files, a plan file to resume, or a plain description; given nothing, it picks the next ready ticket. It checks that the working tree is clean and the branch fits, and offers to split off extra independently shippable goals into deferred-work.md. It investigates the code, then writes a plan file. A change of about 100 lines or fewer gets a short plan and is built in the session. A larger one gets the intent, a code map of what to touch, reuse and leave alone, tasks, Given/When/Then acceptance criteria and verification commands; once you approve, a fresh subagent implements it from the plan alone. Then a reviewer subagent reads the diff; a thorough review runs four lenses instead. The agent checks each finding against the code, fixes simple ones this change caused, defers older problems and logs every verdict in the plan; on the larger route, a finding traced to the plan or the intent reverts the code and re-plans. It then commits and offers a pull request or bmad-walkthrough.

Summarizes the skill at e734d3a.

npx skills add https://github.com/bmad-code-org/bmad-method --skill bmad-build

skills.sh page Source on GitHub All skills from bmad-code-org/bmad-method

Total Downloads: 15,600 on skills.sh, Nov 9, 2026
Repo stars: 55,140 bmad-code-org/bmad-method
First seen: Jul 30, 2026 on skills.sh

Weekly Downloads

1.6K

+58% over the 8 weeks to Nov 9

Weekly values

Week to	Installs
Sep 21	982
Sep 28	1,335
Oct 5	2,672
Oct 12	2,938
Oct 19	1,628
Oct 26	1,753
Nov 2	1,494
Nov 9	1,556

Category	category:engineering category:planning category:debugging category:software-engineering Test-first work, implementation from specs, git hygiene and scope control. Make the agent question you and write the plan down before it writes code. Find root causes, review code and test what you built. Everything for building software: planning, coding, debugging, design, frameworks, cloud, agents and skills.
Type	type:workflow A multi-step process the agent follows: interview, debug, run a TDD loop, review.
Effect	effect:local Its own steps create or change files, git state or local tools.
Needs	needs:cli needs:subagents A specific command-line tool or desktop app must be installed, such as gh, agent-browser, LibreOffice, Obsidian or a vendor CLI. Relies on the agent being able to spawn subagents.
Publisher	by:community Everyone else. Unverified vendor claims default to community.
Topics	implementation planning code-review scope-control agent-workflows

SKILL.md

From bmad-code-org/bmad-method at e734d3a. Checked daily, last changed Nov 6, 2026.

Frontmatter

name: bmad-build
description: 'Implements a story, feature, or bug fix end to end: plan, code, review, verify. Use when the user hands one over and leaves the edits to you; a bare ticket id or link counts. Skip obvious, low-risk mechanical maintenance such as small ignore-file, typo-only, formatting-only, or configuration-hygiene edits. Explicit BMAD requests always qualify. Do not volunteer for user-directed interactive edits or version-control operations that only record existing work.'

Run the following command exactly once without changing the current working directory.

Replace {project-root} with the nearest folder containing _bmad/, from the working directory upward. Replace {skill-root} with the absolute path to this skill's directory.

uv run --no-cache "{project-root}/_bmad/scripts/render_skill.py" --project-root "{project-root}" --skill "{skill-root}"

When the invocation names a route (oneshot or full), append --set workflow.route=<value> to the command.
When the invocation names a review selection (none, quick, or thorough; "skip review" or "no review" mean none), append --set workflow.review=<value> to the command.
The command prints the skill's workflow: follow it. HALT: <reason>: report the reason and stop.
If the script does not exist, BMad is not set up in this project yet. Offer to set it up using bmad skill, then run the command again. If you do not have the bmad skill, offer to install it first.
On any other output or failure, including uv being unavailable, report the command output and stop. Do not run any workflow source directly.