discovery

Guides iterative research with vault context and provenance tracking.

11|1|Updated Mar 28, 2026
One-click install
npx skills add https://github.com/robinslange/learning-loop --skill discovery-robinslange
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: discovery
Source: https://github.com/robinslange/learning-loop/tree/main/skills/discovery
Command: npx skills add https://github.com/robinslange/learning-loop --skill discovery-robinslange

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Interactive topic exploration often lacks structure and provenance; Discovery provides guided, iterative research with vault-backed context and automatic provenance tracking.

Core Features & Use Cases

  • Interactive, style-driven journeys (guided, branch, checkpoint) that steer exploration while the subagents ground findings in your vault.
  • Provenance emission, per-round notes, and optional artifact creation or surf mode to keep the vault organized and actionable.
  • Web-sourced landscape context combined with vault knowledge to surface gaps and actionable insights while preserving source links.

Quick Start

Begin a topic exploration by invoking /discovery with a topic, e.g., /discovery "spaced repetition" to start a guided, vault-backed research journey.

Frequently Asked Questions about discovery

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

FAQPage Schema
How do I guide iterative topic exploration with provenance tracking?

Iterative topic exploration with provenance tracking is guided by invoking a command with a topic, which launches subagents to ground findings in your vault and emit per-round source links. This approach combines web context with vault knowledge to surface actionable insights.

What is vault-backed research and how does it work?

Vault-backed research works by launching subagents that source landscape context from the web and combine it with your existing vault knowledge. This process grounds findings in your personal knowledge base while preserving source links for automatic provenance tracking during interactive topic exploration.

How do I start a guided research journey using vault context?

Start a guided research journey by invoking the discovery command with a chosen topic. The Skill supports warm or cold starts and parameterized styles, allowing you to immediately begin iterative exploration with subagents that produce per-round findings and provenance.

Can I customize the style and tone of my topic exploration?

Yes, you can customize topic exploration using parameterized styles and tones. The Skill supports guided, branch, and checkpoint journeys, and allows you to choose between artifact creation or surf mode to keep your vault organized and actionable based on your preference.

When do I need automatic provenance emission during research?

Automatic provenance emission is needed when you want to trace the origins of your research findings without manual tracking. The Skill emits per-round notes and source links automatically, ensuring vault context remains organized and insights are actionable and verifiable.

Best way to surface knowledge gaps using vault context and web sources?

The best way to surface knowledge gaps is by combining web-sourced landscape context with your vault knowledge through iterative research. Subagents analyze both sources to identify missing information, producing per-round findings that highlight actionable insights while preserving source links.