technical-decision-making

Guides engineering teams through structured technical decisions using reversibility analysis, weighted matrices, and ADRs.

Updated Dec 29, 2025
One-click install
npx skills add https://github.com/snoodleboot-io/discrecontinual_equations --skill technical-decision-making-snoodleboot-io
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: technical-decision-making
Source: https://github.com/snoodleboot-io/discrecontinual_equations/tree/main/.claude/skills/technical-decision-making
Command: npx skills add https://github.com/snoodleboot-io/discrecontinual_equations --skill technical-decision-making-snoodleboot-io

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Engineering teams often stall on technical decisions, over-analyze trivial choices, or commit to irreversible architectures without evidence. This Skill provides a structured framework for sizing decision effort, running time-boxed spikes, and recording decisions so teams decide faster and revisit choices with evidence instead of opinion. ## Core Features & Use Cases - Reversibility Classification: Sort decisions into one-way and two-way doors to calibrate how much analysis each deserves. - Weighted Decision Matrix: Compare options like Celery, SQS, Postgres queues, and Kafka with pre-agreed weights and defined scoring scales. - Spike Design and Revisit Triggers: Define time-boxed experiments with decision rules, plus falsifiers wired to dashboards and alerts. - Use Case: A four-engineer team choosing a background job system scores Postgres queue, SQS, Celery, and Kafka against weighted criteria, records dissent in an ADR, and sets measurable triggers for revisiting the choice. ## Quick Start Ask the AI to help you evaluate a technical decision, such as choosing between two job queue technologies, using the reversibility and weighted matrix framework.

Frequently Asked Questions about technical-decision-making

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

FAQPage Schema
How do I decide between two technologies for my engineering team?▼

Classify the decision as a one-way or two-way door first, then build a weighted decision matrix with criteria agreed before scoring. Score each option including the do-nothing incumbent, and treat results within 0.3 points as tied rather than a clear winner.

What is a one-way door versus two-way door decision?▼

A two-way door is reversible in days to weeks, like a logging library or CI provider, and deserves hours of analysis. A one-way door, like a primary datastore or public API contract, is costly to reverse and warrants spikes, an ADR, and a named owner.

How do I run a technical spike before choosing a technology?▼

Define three things before writing code: the specific question, a time-box, and a decision rule with measurable thresholds. Record the result even when it changes nothing, and avoid spikes when the uncertainty is about preference rather than fact.

When should I write an architecture decision record?▼

Write an ADR for one-way door decisions such as datastores, public APIs, or vendor choices, including alternatives considered, consequences, and recorded dissent. Two-way door decisions only need a PR description.

What are common mistakes in technical decision making?▼

Common anti-patterns include reverse-fitting matrix weights to a preferred option, running unbounded spikes without decision rules, requiring consensus, and skipping revisit triggers. Fix weights before scoring and write falsifiers at decision time.