obsidian-plugin/narthur_research-quest
1
0
Fork 0
mirror of https://github.com/narthur/research-quest.git synced 2026-07-22 11:40:25 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
narthur_research-quest/knowledge.md

2.4 KiB
Raw Permalink Blame History

Research Quest Knowledge Base

Project Overview

An Obsidian plugin that helps users deepen their research by generating intelligent questions and tracking their progress in answering them.

Architecture

  • src/services/ - Core services (OpenAI, storage, events)
  • src/views/ - UI components and views
  • src/__tests__/ - Test suites

Key Concepts

Quest Management

  • Each quest is associated with a specific document
  • Maintain 5 active quests per document
  • Auto-generate new quests when existing ones are completed
  • Track completion status and timestamps

OpenAI Integration

  • Use function calling for reliable structured responses
  • Strict evaluation criteria for marking questions as answered
  • Conservative approach to completion - only mark as done when thoroughly answered

State Management

  • Use event system for cross-component communication
  • Persist quests in Obsidian's data storage
  • Reactive updates when document changes

Development Guidelines

  • Write tests for core functionality
  • Use TypeScript for type safety
  • Follow Obsidian's UI patterns and styles
  • Keep components focused and single-purpose
  • Use Svelte for UI components
  • Conform to Obsidian's plugin guidelines

Testing

  • Run tests before committing: pnpm test
  • Mock external services in tests
  • Test error cases and edge conditions
  • Ensure proper cleanup in tests
  • Run a single test for better debugging: pnpm dlx vitest run -t "should follow the basic flow" src/lib/refreshQuests.spec.ts

Resources

Future Enhancement Ideas

  • Progress tracking and visualization
  • Question organization with tags and grouping
  • Export/import functionality for sharing
  • Enhanced AI features (gap analysis, cross-reference)
  • Collaboration features for team research
  • Deeper Obsidian integration (graph view, links)
  • Research workflow tools
  • UI improvements (drag-drop, search)
  • Smart features (outdated detection, dependencies)
  • Quality control system

Obsolescence Detection Pattern

Track question relevance by:

  • Store content hash with each question
  • Keep minimal context snapshot
  • Regular validation checks
  • AI-assisted relevance evaluation

When obsolete:

  1. Flag for review
  2. Offer regeneration
  3. Move to review queue
  4. Allow manual revalidation