weigh-development-paths

Evaluates competing development approaches across seven dimensions and recommends one with tradeoffs.

Updated May 3, 2026
One-click install
npx skills add https://github.com/Claudfather/clauDNA --skill weigh-development-paths-claudfather
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: weigh-development-paths
Source: https://github.com/Claudfather/clauDNA/tree/main/skills/weigh-development-paths
Command: npx skills add https://github.com/Claudfather/clauDNA --skill weigh-development-paths-claudfather

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When you hit a development junction with multiple viable approaches — architecture choices, refactoring strategies, or where to place new code — picking the wrong option creates rework. This Skill enforces a structured, unbiased comparison before committing to a path. ## Core Features & Use Cases - Seven-Dimension Comparison Matrix: Evaluates every option against elegance, existing patterns, extension, DRY, separation of concerns, future-proofing, and plan alignment, with no blank cells allowed. - Interactive and Autonomous Modes: Runs interactively in plan mode for live decisions, or via --auto with a context bundle as the synthesis pass for automated build pipelines. - Flexible Output Targets: Publishes the junction analysis to chat, GitHub Issues, or a decision-record doc via the shared publish workflow. - Use Case: While implementing a feature you wonder whether to extend an existing module or create a new one. Invoke the skill to get a filled comparison matrix, a holistic assessment of tech-debt trajectory, and a recommendation that names what you lose by not picking the runners-up. ## Quick Start Ask the assistant to weigh whether the new caching logic should live in the existing service module or a new dedicated module, and request the full comparison matrix.

Frequently Asked Questions about weigh-development-paths

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

FAQPage Schema
How do I decide between two architecture approaches in a codebase?

Frame the junction, list all options, then evaluate each against seven dimensions: elegance, existing patterns, extension, DRY, separation of concerns, future-proofing, and plan alignment. Fill the full comparison matrix before synthesizing a recommendation with named tradeoffs.

How to run weigh-development-paths non-interactively in a pipeline?

Invoke it with the --auto flag and a path to a context bundle containing open decisions, codebase-comparison artifacts, and the plan. It emits a structured JSON result with a refined plan, resolved and unresolved decisions, and per-dimension rationales.

Can the junction analysis be published as a GitHub Issue?

Yes. Pass --output github and the skill writes the analysis as a doc and delegates to the publish command, which creates one GitHub Issue containing the comparison matrix, holistic assessment, and recommendation, labeled auto-audit and enhancement.

When should I not use a structured decision matrix?

Skip it when one option is clearly correct, when the decision is trivial such as naming or formatting, or when the analysis is stalling actual building. The matrix is for genuine junctures where the wrong choice creates rework.

What happens when a decision cannot be resolved technically?

In --auto mode, junctions hinging on human values, policy, security, or cost go to decisions_unresolved rather than forcing a pick. The outcome becomes needs-input with the unresolved junctions listed as the blocker description.