draft-enrichment-proposals

Convert taxonomy-classified context-diff findings into structured change proposals and ADR drafts.

3|Updated Jan 25, 2026
One-click install
npx skills add https://github.com/kapilvirenahuja/garura --skill draft-enrichment-proposals
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: draft-enrichment-proposals
Source: https://github.com/kapilvirenahuja/garura/tree/main/core/components/skills/draft-enrichment-proposals
Command: npx skills add https://github.com/kapilvirenahuja/garura --skill draft-enrichment-proposals

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Drafts actionable, reviewer-ready proposals from tiered taxonomy-classified findings so teams can reconcile what should change without re-classifying or writing LTM prematurely.

Core Features & Use Cases

  • Proposal drafting from findings: Generates one proposal per finding, carrying forward the full learning taxonomy exactly as assigned upstream.
  • Target selection for Tier 1–3: Maps each finding to the correct LTM target artifact (or a new file path) and prepares the appropriate action (modify/add/contradict with ADR).
  • Tier 1 ADR drafting: Produces ADR draft(s) using the provided ADR template for Tier 1 items, while keeping LTM writes out of scope.

Quick Start

Use the draft-enrichment-proposals skill to convert diff-context-baseline output at context_diff_path into proposals.yaml and any required ADR drafts at output_base.

Frequently Asked Questions about draft-enrichment-proposals

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

FAQPage Schema
How do I convert context diffs into review-ready change proposals?

Converting context diffs into change proposals involves processing taxonomy-classified findings from a context_diff_path and generating a proposals.yaml artifact. The tool maps each finding to an LTM target artifact, preparing actions like modify or add for team reconciliation.

What is the best way to draft architecture decision records from LTM findings?

Drafting architecture decision records from LTM findings is handled for Tier 1 items using a provided ADR template. The Skill produces ADR draft markdowns while explicitly keeping LTM writes out of scope to prevent premature updates.

How does taxonomy reconciliation work for enterprise software development?

Taxonomy reconciliation for enterprise software development works by validating taxonomy-proposed justification completeness at the proposals boundary. It applies across Tier 1, Tier 2, and Tier 3 scenarios where product or core LTM roots exist for target identification.

Can I use this to map findings to new file paths when LTM roots are missing?

Yes, you can map findings to new file paths when existing LTM roots are missing. The tool supports Tier 3 reconciliation scenarios by allowing you to define new paths for target identification instead of relying on existing artifacts.

When should I not use automated proposal drafting for context diffs?

You should not use automated proposal drafting when you need to immediately write LTM updates, as this tool explicitly defers LTM writes. It is designed only for preparing reviewer-ready proposals and ADR drafts, not executing the actual reconciliation.