One-click install
npx skills add https://github.com/me2resh/apexyard --skill decide-me2resh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: decide
Source: https://github.com/me2resh/apexyard/tree/main/.claude/skills/decide
Command: npx skills add https://github.com/me2resh/apexyard --skill decide-me2resh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents untracked, ad-hoc architecture choices by forcing structured technical decision-making and producing an Agent Decision Record (AgDR) every time.

Core Features & Use Cases

  • Structured decision workflow: Parses the decision topic, gathers only decision-relevant context, and requires listing 2–4 options with pros and cons.
  • Role-gated accountability: Activates the Tech Lead for technical decisions, escalates to Head of Engineering for major architectural thresholds, and includes Security Auditor involvement when auth/crypto/secrets/PII are involved.
  • AgDR creation with portfolio-resolved templates: Generates an auditable AgDR file under the current project’s docs/agdr/ using an adopters’ overridden template when present.
  • Required decision quality rules: Enforces an options table, justification with a mandatory “because” clause, precise ISO-8601 timestamp, and slugging from the title.

Quick Start

Ask for a decision like: decide between deploying a new service as a standalone Lambda or as part of an existing service, considering scalability, cost, and deployment complexity.

Frequently Asked Questions about decide

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

FAQPage Schema
How do I document technical decisions to prevent ad-hoc architecture choices?

To document technical decisions and prevent ad-hoc architecture choices, you can enforce structured technical decision-making that generates an auditable Agent Decision Record. This requires listing 2–4 options with pros and cons and a mandatory justification clause.

What is an Agent Decision Record and when do I need one for architecture governance?

An Agent Decision Record is an auditable file created under a project's docs/agdr folder to track technical decisions. You need one for new services, tech stack changes, integrations, major data model changes, and security-sensitive decisions to ensure architecture governance.

How do I enforce role-based review for technical decisions involving auth or PII?

To enforce role-based review for technical decisions involving auth or PII, you activate a Tech Lead, escalate to a Head of Engineering for major thresholds, and require a Security Auditor sign-off. This role-gated accountability ensures secure architecture governance.

Can I use custom templates for technical decision records in my project?

Yes, you can use custom templates for technical decision records. The system generates the auditable AgDR file using a portfolio-resolved template, automatically applying an adopter's overridden template when one is present in the project.

What is the best way to structure options and justification for a new tech stack element?

The best way to structure options and justification for a new tech stack element is to enforce an options table with 2–4 choices, precise ISO-8601 timestamps, title-based slugging, and a mandatory 'because' clause for justification.