adr-decision-extraction

Extract architectural decisions with problem-solution pairs and trade-offs from conversational text.

Updated Feb 17, 2026
One-click install
npx skills add https://github.com/javierhbr/random-poc --skill adr-decision-extraction
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: adr-decision-extraction
Source: https://github.com/javierhbr/random-poc/tree/main/custom-skills/beagle-main/plugins/beagle-analysis/skills/adr-decision-extraction
Command: npx skills add https://github.com/javierhbr/random-poc --skill adr-decision-extraction

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill automates the extraction of key architectural decisions from conversations, streamlining the process of generating Architecture Decision Records (ADRs).

Core Features & Use Cases

  • Identify Explicit Decisions: Extracts text marked with [ADR] tags.
  • Detect Implicit Choices: Recognizes patterns indicating decisions, trade-offs, and problem-solution pairs.
  • Confidence Scoring: Assigns confidence levels (high, medium, low) to detected decisions.
  • Structured Output: Formats extracted decisions into a JSON object with fields like title, problem, chosen_option, and confidence.
  • Use Case: After a technical design meeting, use this Skill to process the transcript and automatically generate a draft list of ADRs, ensuring important decisions are not lost.

Quick Start

Use the adr-decision-extraction skill to extract architectural decisions from the provided meeting transcript.

Frequently Asked Questions about adr-decision-extraction

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

FAQPage Schema
How do I extract architectural decisions from meeting transcripts?

To extract architectural decisions from meeting transcripts, you can use pattern matching and natural language understanding to identify explicit markers, choice patterns, and problem-solution pairs. This process automates the detection of trade-offs and choices made during technical design discussions.

What is an Architecture Decision Record and how does it capture trade-offs?

An Architecture Decision Record (ADR) captures architectural decisions by documenting the problem, chosen option, and trade-offs. Extracting these records from chat logs or design discussions ensures that contextual inferences and explicit choices are structured into a readable format.

Can I generate ADRs automatically from chat logs and design discussions?

Yes, you can generate ADRs automatically from chat logs and design discussions by applying natural language processing to detect implicit choices and explicit decision markers. The output is structured into a JSON object containing fields like title, problem, and chosen option.

Does this method require specific tags to identify decisions in conversational text?

This method does not strictly require specific tags, as it detects implicit choices through pattern matching and contextual inferences. However, it also identifies explicit decisions marked with [ADR] tags in conversational text to improve extraction accuracy.

How are extracted architectural decisions structured for documentation?

Extracted architectural decisions are structured into a JSON object containing fields like title, problem, chosen option, and confidence. Confidence scoring assigns high, medium, or low levels to detected decisions, ensuring clear documentation for downstream ADR generation.