draft-distill-proposals

Analyze merged PR diffs and issue STM evidence to generate up to two staged proposals.yaml entries.

3|Updated Jan 25, 2026

Search Tags:#learning extraction #STM #evidence analysis #pr diff #deterministic development #proposal staging #enterprise agentic workflow

Ready to run

npx skills add https://github.com/kapilvirenahuja/garura --skill draft-distill-proposals

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: draft-distill-proposals
Source: https://github.com/kapilvirenahuja/garura/tree/main/core/components/skills/draft-distill-proposals
Command: npx skills add https://github.com/kapilvirenahuja/garura --skill draft-distill-proposals

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It analyzes a merged PR diff (optionally with related STM evidence) to detect real learning signals, then produces lightweight, staged proposals so future work can avoid repeating mistakes.

Core Features & Use Cases

Diff-to-proposal learning extraction: Detects semantic learning signals from implementation changes and classifies them into targeted updates.
STM-evidence-aware confidence: Reads available issue evidence (enhance/fix-it artifacts) to raise or lower confidence and prioritize the strongest learnings.
Staged proposals for review: Writes proposals.yaml only to the issue’s STM evidence/distill area, enabling optional human review without updating product LTM.

Quick Start

Use the draft-distill-proposals skill to analyze a merged PR diff along with any available issue STM evidence and generate up to two staged proposals in proposals.yaml for human review.

Frequently Asked Questions about draft-distill-proposals

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

FAQPage Schema

How do I extract learning signals from a merged PR diff?▼

Extracting learning signals from a merged PR diff involves analyzing implementation changes alongside optional issue STM evidence to detect targeted updates. This process generates staged proposals for future review.

What is proposal staging in enterprise agentic workflows?▼

Proposal staging in enterprise agentic workflows is the process of writing detected learning signals to an issue STM evidence area. This keeps product LTM read-only while allowing optional human review before final capture.

How do I generate a proposals.yaml schema from PR evidence?▼

Generating a proposals.yaml schema from PR evidence requires analyzing merged diffs and STM artifacts to classify learning signals. The output is a forward-compatible file capped at two proposals to avoid trivial noise.

Does deterministic development require STM evidence for learning extraction?▼

Deterministic development does not strictly require STM evidence for learning extraction, but reading available issue evidence raises or lowers confidence. This helps prioritize the strongest learnings from the PR diff.

What are the limitations of automated PR diff learning extraction?▼

Limitations of automated PR diff learning extraction include capping output at exactly two staged proposals and avoiding trivial-noise outputs. The system maintains a read-only product LTM to prevent unauthorized updates.

Can I use issue STM artifacts to raise confidence in staged proposals?▼

You can use issue STM artifacts to raise or lower confidence in staged proposals by providing enhance and fix-it evidence. This STM-evidence-aware approach prioritizes the strongest learnings from the diff.