review-decisions

Evaluate decisions in DECISIONS.md using logs, commits, and usage data.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/fernando-fernandez3/claude-code-resources --skill review-decisions
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: review-decisions
Source: https://github.com/fernando-fernandez3/claude-code-resources/tree/main/my-config/skills/review-decisions
Command: npx skills add https://github.com/fernando-fernandez3/claude-code-resources --skill review-decisions

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The DECISIONS.md review process often stalls when decisions flagged for revisiting accumulate data without a clear, repeatable evaluation method. This Skill provides a structured, data-driven approach to audit, update, and summarize decisions, reducing manual drift and speeding up governance.

Core Features & Use Cases

  • Automated evaluation: Reads DECISIONS.md, gathers metrics from logs, commits, and usage data, and generates clear recommendations.
  • Decision lifecycle management: Updates DECISIONS.md to mark Active, Pending, or Superseded decisions with traceable context.
  • Review summaries: Produces a concise review entry with data-supported insights for leadership or collaboration.

Quick Start

  1. Run the review-decisions skill to assess decisions currently marked for revisit in DECISIONS.md.
  2. Review the generated recommendations and update the DECISIONS.md accordingly.
  3. If needed, create a new pending decision or supersede an old one with linked data.

Frequently Asked Questions about review-decisions

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

FAQPage Schema
How do I automate the review of decisions logged in DECISIONS.md?

Automating DECISIONS.md reviews involves reading the file, aggregating metrics from logs and commits, and generating actionable recommendations to update decision statuses with traceable context.

What is data-driven governance for decision lifecycle management?

Data-driven governance evaluates decisions marked for revisit by collecting usage data and commit metrics, then updates their status to Active, Pending, or Superseded with concise, data-supported summaries.

How do I update decision statuses in DECISIONS.md based on usage metrics?

Updating decision statuses requires aggregating commit and log data to evaluate flagged decisions, then marking them Active, Pending, or Superseded while appending a review summary with linked metrics.

Can I generate review summaries for leadership from accumulated decision logs?

Yes, generating review summaries for leadership involves aggregating data from logs and commits to produce concise entries with data-supported insights and structured next steps for collaboration.

What is the best way to audit pending decisions that have accumulated data?

Auditing pending decisions requires a structured approach that reads DECISIONS.md, aggregates usage metrics, and produces clear recommendations to reduce manual drift and speed up governance.

When should I supersede an old decision in DECISIONS.md?

You should supersede an old decision when aggregated metrics and usage data indicate it is outdated, creating a new pending decision with linked data and traceable context for governance.