structured-doc-authoring

Author structured PRDs, design docs, plans, ADRs, roadmaps, and RCAs with templates and diagrams.

Updated Jul 8, 2026
One-click install
npx skills add https://github.com/Zhachory1/autopraxis --skill structured-doc-authoring-zhachory1
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: structured-doc-authoring
Source: https://github.com/Zhachory1/autopraxis/tree/main/skills/structured-doc-authoring
Command: npx skills add https://github.com/Zhachory1/autopraxis --skill structured-doc-authoring-zhachory1

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Writing decision-grade workflow documents (PRDs, design docs, implementation plans, ADRs, roadmaps, RCAs) is slow and inconsistent, often producing walls of text that downstream agents and reviewers cannot act on. This Skill standardizes doc authoring with fixed templates, evidence standards, and required diagrams so documents are reviewable and executable. ## Core Features & Use Cases - Template-driven authoring: Six ready templates covering PRD, design doc/RFC, technical plan, ADR, roadmap, and RCA, each with fixed sections for goals, non-goals, tradeoffs, and acceptance criteria. - Diagram-first documentation: Enforces Mermaid and Graphviz/DOT diagrams for flows, dependencies, timelines, and causal chains, with required interpretation notes. - Quality gates and standards: A standards reference defines document selection, SPADE decision framing, evidence confidence labels, review checklists, and anti-patterns. - Use Case: After an incident, ask the agent to produce an RCA. It loads the RCA template, builds a timeline and causal Graphviz diagram, records confirmed root cause and ruled-out hypotheses, and lists prevention actions with owners. ## Quick Start Use the structured-doc-authoring skill to write a design doc for the new caching layer, including a Mermaid sequence diagram, alternatives considered, and a rollout plan.

Frequently Asked Questions about structured-doc-authoring

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

FAQPage Schema
How do I write a PRD with an AI agent?▼

Provide the problem, target user, success metric, and constraints, then the agent fills the PRD template covering problem alignment, goals, non-goals, key flows, launch plan, and open questions. It adds a Mermaid user-journey diagram and a review state table for approvers.

What is the difference between a design doc and a technical plan?▼

A design doc explains how to build something: architecture, alternatives, tradeoffs, and cross-cutting concerns. A technical plan comes after design approval and sequences implementation into tasks with file maps, acceptance criteria, and validation commands.

When should I use Mermaid versus Graphviz for diagrams?▼

Use Mermaid for flows, sequences, timelines, gantts, and state machines that must render in GitHub Markdown. Use Graphviz/DOT for dense dependency graphs, decision branches, and RCA causal chains where layout control matters more than native rendering.

How do I write an RCA that goes beyond the proximate cause?▼

The RCA template requires a confirmed root cause with evidence, contributing system factors, ruled-out hypotheses, and prevention actions with owners and success evidence. It uses a causal Graphviz diagram separating trigger, latent condition, failure mode, and detection gap.

What is a SPADE decision framework in a design doc?▼

SPADE structures hard decisions into Setting, People, Alternatives, Decision, and Explanation. The templates include a SPADE block naming the responsible owner, consulted stakeholders, feasible alternatives with pros and cons, and the communication plan.

When should I not use a full document template?▼

For trivial documents, use lightweight inline sections instead of full templates. Full templates with diagrams and SPADE framing are reserved for non-trivial PRDs, DDs, plans, ADRs, roadmaps, and RCAs where decisions carry real risk.