kata-discovery-synthesis

Synthesize heterogeneous research sources into structured discovery insight markdown files.

Updated Sep 3, 2025
One-click install
npx skills add https://github.com/guardiatechnology/design-system --skill kata-discovery-synthesis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kata-discovery-synthesis
Source: https://github.com/guardiatechnology/design-system/tree/main/.claude/skills/kata-discovery-synthesis
Command: npx skills add https://github.com/guardiatechnology/design-system --skill kata-discovery-synthesis

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps teams replace manual reading and note-taking across heterogeneous sources with consistent, structured “discovery insights” that are ready for review.

Core Features & Use Cases

  • Guided source ingestion: Reads heterogeneous inputs (e.g., Notion, Figma, GitHub, or local transcripts) and captures evidence.
  • Insight candidate extraction: Breaks observations into indivisible, traceable units and avoids mixing multiple pains into one insight.
  • Canonical insight generation: Produces an insight markdown file with required front-matter and a standardized body (Observation, Source, Initial implication, Open questions) suitable for docs/discovery/{topic}/insights/.

Quick Start

Ask the AI to synthesize discovery insights for topic scheduled-payments-research from the provided source_refs, creating the next available numbered insight files in docs/discovery/scheduled-payments-research/insights/.

Frequently Asked Questions about kata-discovery-synthesis

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

FAQPage Schema
How do I synthesize product discovery insights from heterogeneous research sources?

Product discovery insight synthesis works by reading heterogeneous inputs and transforming each observation into a single traceable insight markdown file under docs/discovery/{topic}/insights/. It enforces traceability via source_refs and uses a canonical front-matter schema to separate Observation, Source, Initial implication, and Open questions without proposing solutions.

What is the best way to structure messy research notes into traceable insights?

The best way to structure messy research notes is to break observations into indivisible units, avoiding mixing multiple pains into one insight. This Skill generates a standardized markdown body with required front-matter, capturing a single observation with source_refs for full traceability across heterogeneous inputs.

Can I ingest source references from tools like Notion, Figma, or GitHub for insight generation?

Yes, guided source ingestion supports reading heterogeneous inputs from Notion, Figma, GitHub, or local transcripts. It captures evidence from these source_refs to generate structured discovery insights ready for team review.

How do I generate markdown files with canonical front-matter for product discovery?

To generate markdown files with canonical front-matter, ask the Skill to synthesize insights for a specific topic from provided source_refs. It automatically creates the next available numbered insight files in the docs/discovery/{topic}/insights/ directory using the required schema.

Does discovery insight synthesis propose solutions for the open questions it identifies?

No, discovery insight synthesis does not propose solutions. It strictly separates Observation, Source, Initial implication, and Open questions to ensure the output remains an objective, traceable insight ready for manual review.