bmad-help

Analyzes BMad workflow state and recommends the next skill to run.

1|Updated Sep 18, 2026
One-click install
npx skills add https://github.com/PastaSus/egg-defender --skill bmad-help-pastasus
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-help
Source: https://github.com/PastaSus/egg-defender/tree/main/.agents/skills/bmad-help
Command: npx skills add https://github.com/PastaSus/egg-defender --skill bmad-help-pastasus

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Users working with the BMad framework often lose track of which module and phase they are in, what steps are already complete, and which skill to invoke next. This Skill reads the installed skill catalog, resolved configuration, and existing output artifacts to orient the user and answer BMad-related questions. ## Core Features & Use Cases - Workflow Position Detection: Scans resolved output locations for artifact files and fuzzy-matches them to catalog rows to infer which steps have started or completed. - Next-Step Recommendation: Uses phase, preceded-by, followed-by, and required fields from the bmad-help.csv catalog to recommend optional and required next skills with menu codes and invocation args. - Grounded Q&A: Fetches module documentation referenced by _meta catalog rows to answer general questions about BMad modules without fabricating project details. - Use Case: A user returns to a project after a week and asks "what should I do next in BMad?" The Skill detects the completed PRD artifact, identifies the next required gate, and offers to launch that skill immediately. ## Quick Start Ask the assistant where you are in your BMad workflow and what skill to run next.

Frequently Asked Questions about bmad-help

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

FAQPage Schema
How do I find out what to do next in BMad?

Invoke the bmad-help skill and ask what to do next. It reads the bmad-help.csv catalog, checks resolved output paths for completed artifacts, and recommends the next optional and required skills with their menu codes and invocation commands.

How does BMad detect which workflow steps are completed?

The skill searches resolved output-location paths for files matching the outputs patterns in the catalog and fuzzy-matches them to skill rows. It inspects artifacts for explicit completion markers, since a draft file only proves a skill started, not that it finished.

Can bmad-help answer general questions about BMad modules?

Yes. Catalog rows with _meta in the skill column carry a URL or path to module documentation. The skill fetches these sources to ground answers about modules when the question does not map to a specific skill.

What configuration does bmad-help need to work?

It runs resolve_config.py with uv to merge _bmad/config.toml, config.user.toml, and custom config files. The merged JSON provides output-location variables, communication language, and the project_knowledge path used for grounding.

What are the limitations of BMad completion detection?

File presence alone does not prove completion, since artifacts may be drafts. When completion cannot be determined reliably, the skill says so and asks the user rather than recommending advancement based on uncertain evidence.