bmad-spec

Distill transcripts, PRDs, and RFCs into a structured five-field kernel with an append-only .memlog.md file.

Updated Jul 11, 2026
One-click install
npx skills add https://github.com/marcmontecalvo/kinward --skill bmad-spec-marcmontecalvo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-spec
Source: https://github.com/marcmontecalvo/kinward/tree/main/.agents/skills/bmad-spec
Command: npx skills add https://github.com/marcmontecalvo/kinward --skill bmad-spec-marcmontecalvo

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires uv, and includes scripts (resource) and assets (resource) components.

What problem does it solve?

This skill solves the problem of fragmented, inconsistent, and unvalidated project requirements by distilling multi-source inputs into a single, canonical machine-readable contract.

Core Features & Use Cases

  • Canonical Distillation: Transforms brain dumps, transcripts, and PRDs into a structured five-field kernel (Why, Capabilities, Constraints, Non-goals, Success signal).
  • Preservation-Validated Memory: Maintains an append-only .memlog.md file that ensures every decision and capability ID remains stable and traceable across updates.
  • Use Case: When a team has a collection of Slack threads, emails, and mockups for a new feature, this skill synthesizes them into a formal SPEC.md that downstream engineering and design agents can consume as a single source of truth.

Quick Start

Ask the bmad-spec skill to create a spec for the new user authentication module based on the provided project context.

Frequently Asked Questions about bmad-spec

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

FAQPage Schema
How do I consolidate fragmented product requirements from transcripts and PRDs into a single source of truth?

To consolidate fragmented product requirements from transcripts and PRDs, you distill multi-source inputs into a structured five-field kernel covering Why, Capabilities, Constraints, Non-goals, and Success signal, producing a canonical machine-readable contract.

What is a machine contract specification and when do I need one for product management?

A machine contract specification is a structured, validated document that distills project intent into a stable kernel and append-only decision log. You need one to maintain traceable requirements when managing complex product workflows across diverse input formats.

How do I maintain stable capability tracking and decision logs when updating project specifications?

To maintain stable capability tracking when updating specifications, you use a preservation-validated append-only memory log file that ensures every decision and capability ID remains stable and traceable across updates.

Do I need uv installed to run the bmad-spec workflow automation for requirements engineering?

Yes, you need the uv dependency installed to run the bmad-spec workflow automation, which processes multi-source inputs into a validated machine contract for requirements engineering.

Can I use Slack threads and emails as input formats for generating a formal specification document?

Yes, you can use Slack threads, emails, and mockups as input formats. The skill synthesizes diverse inputs like transcripts and PRDs into a formal SPEC.md that downstream agents can consume as a single source of truth.

What are the limitations of using a spec-kernel ecosystem for requirements engineering?

A limitation of using a spec-kernel ecosystem is the strict adherence required to its rules for capability tracking and load-bearing content management, meaning inputs must fit the five-field kernel structure of Why, Capabilities, Constraints, Non-goals, and Success signal.