bmad-workflow-builder
Builds, converts, and analyzes workflows and skills. Use when the user requests to "build a workflow", "modify a workflow", "quality check workflow", "analyze skill", or "convert a skill".
- License 许可
- MIT
- 最近更新
- 2026-04-18
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- bmad-workflow-builder
- description
- Builds, converts, and analyzes workflows and skills. Use when the user requests to "build a workflow", "modify a workflow", "quality check workflow", "analyze skill", or "convert a skill".
02 资源文件 27
- 资产文件assets/SKILL-template.md1.3 KB
- 参考文档references/build-process.md9.3 KB
- 参考文档references/classification-reference.md3.5 KB
- 参考文档references/complex-workflow-patterns.md4.3 KB
- 参考文档references/convert-process.md5.4 KB
- 参考文档references/quality-analysis.md7.4 KB
- 参考文档references/quality-dimensions.md3.1 KB
- 参考文档references/quality-scan-enhancement-opportunities.md15.6 KB
- 参考文档references/quality-scan-execution-efficiency.md15.1 KB
- 参考文档references/quality-scan-prompt-craft.md28.3 KB
- 参考文档references/quality-scan-script-opportunities.md10.8 KB
- 参考文档references/quality-scan-skill-cohesion.md9.9 KB
- 参考文档references/quality-scan-workflow-integrity.md17.8 KB
- 参考文档references/report-quality-scan-creator.md8.9 KB
- 参考文档references/script-opportunities-reference.md4.7 KB
- 参考文档references/script-standards.md4.4 KB
- 参考文档references/skill-best-practices.md8.0 KB
- 参考文档references/standard-fields.md6.9 KB
- 参考文档references/template-substitution-rules.md1.8 KB
- 脚本scripts/generate-convert-report.py15.9 KB
- 脚本scripts/generate-html-report.py23.9 KB
- 脚本scripts/prepass-execution-deps.py9.3 KB
- 脚本scripts/prepass-prompt-metrics.py10.4 KB
- 脚本scripts/prepass-workflow-integrity.py16.3 KB
- 脚本scripts/scan-path-standards.py10.6 KB
- 脚本scripts/scan-scripts.py26.2 KB
- 脚本scripts/tests/test_generate_convert_report.py9.3 KB
03 SKILL.md 正文
Workflow & Skill Builder
Overview
This skill helps you build AI workflows and skills that are outcome-driven — describing what to achieve, not micromanaging how to get there. LLMs are powerful reasoners. Great skills give them mission context and desired outcomes; poor skills drown them in mechanical procedures they'd figure out naturally. Your job is to help users articulate the outcomes they want, then build the leanest possible skill that delivers them.
Act as an architect guide — walk users through conversational discovery to understand their vision, then craft skill structures that trust the executing LLM's judgment. The best skill is the one where every instruction carries its weight and nothing tells the LLM how to do what it already knows.
Args: Accepts --headless / -H for non-interactive execution, --convert <path-or-url> to convert an existing skill into a lean equivalent with before/after HTML comparison report, an initial description for create, or a path to an existing skill with keywords like analyze, edit, or rebuild.
Your output: A skill structure ready to integrate into a module or use standalone — from simple composable utilities to complex multi-stage workflows.
On Activation
-
Detect user's intent. If
--headlessor-His passed, or intent is clearly non-interactive, set{headless_mode}=truefor all sub-prompts. -
Load available config from
{project-root}/_bmad/config.yamland{project-root}/_bmad/config.user.yaml(root and bmb section). If missing, and thebmad-builder-setupskill is available, let the user know they can run it at any time to configure. Resolve and apply throughout the session (defaults in parens):{user_name}(default: null) — address the user by name{communication_language}(default: user or system intent) — use for all communications{document_output_language}(default: user or system intent) — use for generated document content{bmad_builder_output_folder}(default:{project-root}/skills) — save built agents here{bmad_builder_reports}(default:{project-root}/skills/reports) — save reports (quality, eval, planning) here
-
Route by intent — see Quick Reference below.
Build Process
The core creative path — where workflow and skill ideas become reality. Through conversational discovery, you guide users from a rough vision to a complete, outcome-driven skill structure. This covers building new skills from scratch, converting non-compliant formats, editing existing ones, and rebuilding from intent.
Load references/build-process.md to begin.
Quality Analysis
Comprehensive quality analysis toward outcome-driven design. Analyzes existing skills for over-specification, structural issues, execution efficiency, and enhancement opportunities. Uses deterministic lint scripts and parallel LLM scanner subagents. Produces a synthesized report with themes and actionable opportunities.
Load references/quality-analysis.md to begin.
Convert
One-command conversion of any existing skill into a BMad-compliant, outcome-driven equivalent. Whether the input is bloated, poorly structured, or just doesn't follow BMad best practices, this path reads or fetches the original, rebuilds from intent (always headless), and generates an HTML comparison report showing the before/after — metrics, what changed and why, what survived and why it earned its place.
--convert implies headless mode. Accepts a local path or URL. The original skill provides all context needed — no interactive discovery.
Load references/convert-process.md to begin.
Skill Intent Routing Reference
| Intent | Trigger Phrases | Route |
|---|---|---|
| Build new | "build/create/design a workflow/skill/tool" | Load references/build-process.md |
| Convert | --convert path-or-url |
Load references/convert-process.md |
| Existing skill provided | Path to existing skill, or "edit/fix/analyze" | Ask the 3-way question below, then route |
| Quality analyze | "quality check", "validate", "review workflow/skill" | Load references/quality-analysis.md |
| Unclear | — | Present options and ask |
When given an existing skill, ask:
- Analyze — Run quality analysis: identify opportunities, prune over-specification, get an actionable report
- Edit — Modify specific behavior while keeping the current approach
- Rebuild — Rethink from core outcomes using this as reference material, full discovery process
Analyze routes to references/quality-analysis.md. Edit and Rebuild both route to references/build-process.md with the chosen intent.
Regardless of path, respect headless mode if requested.
04 同仓库更多 Skills
meeting-minutes
Generate concise, actionable meeting minutes for internal meetings. Includes metadata, attendees, agenda, decisions, action items (owner + due date), and follow-up steps.
documentation-writer
Diátaxis Documentation Expert. An expert technical writer specializing in creating high-quality software documentation, guided by the principles and structure of the Diátaxis technical documentation authoring framework.
doc-coauthoring
Guide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks.
code-review-workflow
Executes comprehensive 4-phase code review: basic review, quality check, bundle analysis, and test coverage. Should be used when the user mentions comprehensive code review, checking code quality, or reviewing staged changes with multi-phase analysis. This is the comprehensive review workflow, distinct from focused skills like frontend-code-review, staged-code-review, or code-change-test-designer.
code-change-test-designer
Analyzes code changes and designs comprehensive test cases. Reads diffs, summarizes capability changes, and designs test cases covering main flows and edge cases. Should be used when the user mentions test coverage, designing tests for code changes, or analyzing diff/commit for testing purposes. Distinguished from code-review-workflow which provides comprehensive review, this skill focuses on test case design for code changes.
skill-creator
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.