bmad-spec

Distill intent inputs into a five-field SPEC.md kernel with companion files.

Updated Jun 3, 2026
One-click install
npx skills add https://github.com/MathematicGuy/team-006-vin20k --skill bmad-spec-mathematicguy
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: bmad-spec
Source: https://github.com/MathematicGuy/team-006-vin20k/tree/main/.agent/skills/bmad-spec
Command: npx skills add https://github.com/MathematicGuy/team-006-vin20k --skill bmad-spec-mathematicguy

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes assets (resource) components.

What problem does it solve? Raw intent inputs like brain dumps, PRDs, meeting transcripts, or customer emails are unstructured and inconsistent, making them unreliable contracts for downstream planning and implementation work. ## Core Features & Use Cases - Spec Distillation: Converts any intent input into a canonical SPEC.md carrying the five-field kernel (Why, Capabilities, Constraints, Non-goals, Success signal) with stable capability IDs. - Companion Management: Routes load-bearing overflow content such as glossaries, diagrams, and entity catalogs into named companion files referenced from SPEC.md frontmatter. - Preservation Validation: Runs a two-pass self-validation sweep checking Spec Law rules and confirming every load-bearing source claim is preserved, with verdicts logged to .decision-log.md. - Use Case: Paste a messy Slack thread describing a new feature and receive a spec folder containing SPEC.md, a glossary companion, and a decision log ready for downstream planning skills. ## Quick Start Ask the agent to distill the attached product brief into a spec using the bmad-spec workflow.

Frequently Asked Questions about bmad-spec

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

FAQPage Schema
How do I turn a PRD or brain dump into a structured spec?▼

Provide the input file or pasted content to the skill, which distills it into a SPEC.md kernel with Why, Capabilities, Constraints, Non-goals, and Success signal fields. Overflow content like glossaries or diagrams is routed into companion files referenced in the frontmatter.

What inputs does the spec distillation skill accept?▼

It accepts vague ideas, brain dumps, PRDs, GDDs, RFCs, briefs, Slack threads, customer emails, meeting transcripts, mockups, and mixed multi-source material. Structured inputs are lifted directly; mixed inputs are sorted claim by claim.

Can I update an existing spec without losing capability IDs?▼

Yes. Invoking the skill with the same slug targets the existing spec folder and updates it in place. Capability IDs are preserved, new capabilities get the next unused CAP-N, and retired IDs are never reused.

What happens when the input is too thin to distill?▼

The skill blocks with an insufficient_intent error code in headless mode or asks for more detail interactively. It suggests using an upstream elicitation skill like bmad-prd instead of coaching the idea out itself.

How does spec validation work after creation?▼

A two-pass sweep runs after every create or update: pass one checks coherence against Spec Law rules, and pass two walks the source claim by claim to confirm preservation. The verdict is appended to .decision-log.md.