bmad-walkthrough

Guides human reviewers through code changes with orientation, walkthrough, risk, and testing steps.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Reviewing a code change without structure leads to missed risks and shallow approvals. This Skill walks a human reviewer through a change step by step, from understanding its intent to making an approve, rework, or discuss decision. ## Core Features & Use Cases - Guided Review Trail: Identifies the change from a PR, commit, branch, or spec, then builds a concern-based walkthrough with clickable path:line stops. - Risk and Testing Passes: Surfaces high-blast-radius spots (auth, schema, billing, security) and suggests manual observations of observable behavior. - Decision Wrap-Up: Prompts for an approve, rework, or discuss decision and can help apply patches or approve a PR via gh. - Use Case: After a teammate opens a PR for an auth refactor, ask for a walkthrough to get the intent summary, a suggested review order, risk spots, and hands-on verification steps before approving. ## Quick Start Ask the assistant to walk you through this change or PR so you can review it step by step.

Frequently Asked Questions about bmad-walkthrough

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

FAQPage Schema
How do I walk through a code change for human review?

Trigger the walkthrough with a PR, commit SHA, branch, or spec file. The skill identifies the change, summarizes its intent, builds a concern-based review trail with path:line stops, then guides you through risk spots, testing suggestions, and a final decision.

How does the walkthrough decide what change to review?

It checks in order: an explicit argument, recent conversation context, sprint status files for stories in review, and the current git HEAD. If none identify a change, it asks you directly and halts after three unresolved exchanges.

Can I use this review workflow without a spec file?

Yes. Without a spec it runs in bare-commit mode, using the commit message as the intent source and generating a review trail from the git diff. Terse commit messages get an inferred intent flagged for your correction.

What happens if git is unavailable during the review?

Trail generation fails gracefully and the workflow falls back to a non-trail path in the walkthrough step. Surface area stats are skipped with a note that they could not be computed.

Can I customize the walkthrough workflow steps?

Yes. A customize.toml file supports prepend and append activation steps, persistent facts, and an on_complete hook, with team and user override files merged using BMad structural rules.