Architecture Decision Record Generator for Openclaw

A specialized tool for analyzing codebases and conversations to document the strategic reasoning behind technical choices.

charlie-morrison
v1.0.1
May 1, 2026
0
792
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install adr-generator

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install adr-generator using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Architecture Decision Record Generator?

The ADR Generator is a technical documentation tool designed to capture the context and rationale behind significant architectural choices. By leveraging this Openclaw Skills capability, development teams can create a permanent record of why specific frameworks, databases, or conventions were chosen, preventing recurring debates and easing the onboarding process for new members.

This skill automates the discovery of historical context through git logs and codebase analysis, ensuring that Architecture Decision Records (ADRs) are grounded in technical truth. It follows the industry-standard Michael Nygard format to maintain consistency and clarity across the documentation lifecycle.

Architecture Decision Record Generator Use Cases

  • Documenting the selection of a new database, framework, or major library.
  • Recording API contracts and data schema definitions for future reference.
  • Formalizing team conventions such as testing strategies or branching models.
  • Capturing trade-offs made during high-pressure development phases.
  • Creating a historical log of tool adoption or deprecation within a project.

How Architecture Decision Record Generator Works

  1. Identify the core decision by extracting key information from current conversations or code reviews.
  2. Reconstruct context by scanning git history and existing documentation for related changes and keywords.
  3. Analyze alternatives by evaluating traces of experiments, spikes, or deleted dependency references.
  4. Assess consequences by searching the codebase for implementation depth, workarounds, or technical debt.
  5. Generate the final ADR using a structured markdown template including status, deciders, and references.

Architecture Decision Record Generator Setup

To initialize the ADR directory and start tracking decisions using Openclaw Skills, use the following commands:


The skill will automatically detect standard paths like `docs/decisions/` or `architecture/` during its analysis phase.

Architecture Decision Record Generator Data Schema & Taxonomy

The ADR Generator organizes data using a sequential, file-based taxonomy to ensure immutability and easy versioning.

Feature Specification
File Format Markdown (.md)
Naming Convention NNNN-short-noun-phrase.md (e.g., 0001-use-postgresql.md)
Location docs/adr/ or doc/architecture/
Index File A README.md containing a tabular summary of all decisions, statuses, and dates.
Template Michael Nygard format (Context, Decision, Alternatives, Consequences).

Architecture Decision Record Generator Advanced Features

  • Automated Git Forensics: Uses advanced git filtering to find when dependencies were added or deleted to reconstruct decision timelines.
  • Heuristic Alternative Discovery: Scans for keywords like 'experiment' or 'poc' in commit messages to find discarded technical paths.
  • Impact Mapping: Quantifies how deeply a choice is embedded by counting framework-specific imports across the codebase.
  • Supersession Tracking: Automatically manages the lifecycle of decisions by linking new ADRs to the records they supersede.
  • Multi-Agent Integration: Can be triggered by other agents during PR reviews or architecture brainstorming sessions.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*