Bmad builder

Skill xmm/codex-bmad-skills/skills/bmad-builder

BMAD skills and workflows for OpenAI Codex (App, CLI, Web): intent-based execution, YAML project state, and reusable skill packs for planning, architecture, sprint delivery, development, and code review.

Install

npx -y skills add xmm/codex-bmad-skills --skill bmad-builder

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.

What its author says it does

Copied from the file, not written here

Skill and workflow scaffolding skill for BMAD. Use for bmad:create-skill and bmad:create-workflow to extend BMAD with domain-specific capabilities.

SKILL.md

3.2 KB, as published. Nobody here has run it

BMAD Builder

Trigger Intents

bmad:create-skill
bmad:create-workflow

Workflow Variants

create-skill

Scaffold a new BMAD skill folder with minimal required files.

create-workflow

Define workflow contract, artifacts, and trigger semantics for a domain task.

Inputs

capability goal and scope
intended triggers/intents
expected inputs, outputs, and quality checks

Language Guard (Mandatory)

Enforce language selection separately for chat responses and generated artifacts.

Chat language (communication_language) fallback order:

language.communication_language from bmad/project.yaml
English

Rules for chat responses:

Use the resolved chat language for all assistant responses (questions, status updates, summaries, and handoff notes).
Do not switch chat language unless the user explicitly requests a different language in the current thread.

Artifact language (document_output_language) fallback order:

language.document_output_language from bmad/project.yaml
English

Rules for generated artifacts:

Use the resolved artifact language for all generated BMAD documents and structured artifacts.
write prose and field values in the resolved document language
avoid mixed-language requirement clauses with English modal verbs (for example, System shall followed by non-English text)
allow English acronyms/abbreviations in non-English sentences (for example, API, SLA, KPI, OAuth, WCAG)
Keep code snippets, CLI commands, file paths, and identifiers in their original technical form.

Mandatory Reference Load

Before executing create-skill or create-workflow, read REFERENCE.md first. Treat REFERENCE.md as required context before applying scaffold patterns.

Output Contract

new skill scaffold under skills/
required files: SKILL.md, agents/openai.yaml
optional files: templates/, scripts/, resources/

Core Workflow

Define role, trigger semantics, and artifact contract.
Scaffold minimal structure.
Add deterministic scripts only where repeatability is required.
Validate frontmatter and trigger clarity.
Provide usage prompt and handoff rules.

Script Selection

Create scaffold:

bash scripts/scaffold-skill.sh <skill-name>

Validate resulting skill:

bash scripts/validate-skill.sh <skill-dir>

Template Map

templates/skill.template.md
Why: baseline structure for new skill instructions.
templates/workflow.template.md
Why: workflow contract template.
templates/document.template.md
Why: reusable artifact template.

Reference Map

REFERENCE.md
Must read first for extended builder guidance and quality checks.
resources/skill-patterns.md
Use for naming, trigger, and structure patterns.

Quality Gates

skill metadata is valid and discoverable
triggers and outputs are unambiguous
generated structure is minimal and maintainable
no runtime-specific legacy dependencies are introduced