adr-curator

Extract and organize ADR candidates from CodeBuddy review artifacts and repository evidence.

4|1|Updated Feb 3, 2026
One-click install
npx skills add https://github.com/danielbaustin/agent-design-language --skill adr-curator
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: adr-curator
Source: https://github.com/danielbaustin/agent-design-language/tree/main/adl/tools/skills/adr-curator
Command: npx skills add https://github.com/danielbaustin/agent-design-language --skill adr-curator

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Draft ADR candidates from scattered evidence without mutating repositories, enabling teams to review decisions before implementation.

Core Features & Use Cases

  • Identify candidate ADRs from CodeBuddy packets, architecture reviews, findings, migration notes, docs, PR notes, and repo evidence.
  • Preserve evidence and capture status (proposed, accepted_existing, superseded_existing, deferred), context, decision, consequences, alternatives, and validation notes.
  • Build a ready-to-review ADR catalog to feed architecture governance, decision-making, or implementation planning.

Quick Start

Run the deterministic ADR curator against your review packets and repo evidence to emit a candidate ADR set.

Frequently Asked Questions about adr-curator

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

FAQPage Schema
How do I generate Architecture Decision Record candidates from existing code evidence?

You can generate Architecture Decision Record candidates by extracting and organizing review artifacts, migration notes, and repository evidence into a structured catalog with status, context, and supersession links.

Can I draft ADRs from CodeBuddy review packets without modifying the repository?

Yes, you can draft ADRs from CodeBuddy review packets without modifying the repository, as the extraction process operates on review packets and repo signals to produce a candidate catalog without mutating customer code.

What is the best way to track architecture decisions and supersession links from code reviews?

The best way to track architecture decisions is to use deterministic extraction on review packets and repo evidence, capturing status tags like proposed, accepted_existing, or superseded_existing to build a ready-to-review ADR catalog.

Does drafting ADR candidates automatically accept or implement the architecture decisions?

No, drafting ADR candidates does not automatically accept or implement decisions, because the process enforces a strict approval boundary that prevents automatic acceptance or repository changes.

What context and metadata are captured when curating ADR candidates from repository evidence?

When curating ADR candidates from repository evidence, the process captures status, context, decision details, consequences, alternatives, and validation notes to support architecture governance and planning.

What sources can I use to identify ADR candidates for architecture governance?

You can identify ADR candidates from CodeBuddy packets, architecture reviews, findings, migration notes, docs, PR notes, and repo evidence to feed architecture governance and decision-making workflows.