When does this skill trigger?

It activates whenever you are about to make an architectural choice, add a new entity, choose a pattern, or resolve a non-obvious technical trade-off.

Why shouldn't I just use standard industry patterns?

Project-specific decisions documented in the log override general defaults to maintain internal consistency and avoid breaking project-specific logic or data structures.

How do I log a new decision with this skill?

After you and Claude resolve a non-obvious trade-off, the skill will automatically offer to document the outcome and rationale in your DECISIONS.md file.

Does this skill slow down development speed?

While it adds a brief verification step, it prevents significant future rework caused by making the wrong technical choices under time pressure.

What is a DECISIONS.md file?

It is a project-level document that records foundational architectural choices, rationale, and technical trade-offs to ensure long-term consistency beyond what is visible in the code.

Decision Log Manager

Name: Decision Log Manager
Author: bryanmig

bybryanmig

0•

コラボレーションツール

Enforces architectural consistency by checking and maintaining project-specific technical decisions in a DECISIONS.md file.

The Decision Log skill ensures that every architectural choice, from data modeling to library selection, respects existing project precedents. It prevents common pitfalls by identifying 'red flags' like non-standard ID formats or premature abstractions, requiring the AI to verify historical rationale before proceeding. By providing a structured workflow for citing past choices and proposing new entries for the log, it helps development teams maintain technical debt control and uphold project-specific standards that standard industry defaults might otherwise overlook.

主な機能

01Mandatory DECISIONS.md verification before architectural changes

02Distinguishes between architectural rationale and simple style preferences

03Automated citation of past technical decisions in code implementations

04Workflow prompts to log new non-obvious technical trade-offs

050 GitHub stars

06Recognition of 'Red Flag' scenarios requiring extra scrutiny

ユースケース

01Onboarding AI agents to projects with established architectural constraints

02Documenting the rationale behind complex technical trade-offs for future maintenance

03Ensuring consistent data modeling and ID structures across a complex codebase

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add bryanmig/claude-plugins decision-log

For use in Claude.ai and ChatGPT

Download Skill