External research conducted for llm-tutor. Each entry credits the sources it leaned on and links forward to where the findings landed (persona files, README sections, forthcoming skills/schemas). See the /flagrare:research-catalog skill for the workflow, or just read an entry to see the shape.
| Date | Topic | Triggered by | Informed |
|---|---|---|---|
| 2026-06-02 | llm-tutor Design Foundations | Pivot to "meta-tutor" vision; needed empirical grounding before /tutor-start and the gamification model |
Persona files (anti-dependency philosophy), README (convivial-tool framing, "teach me this codebase" positioning), forthcoming /tutor-start and state.json |
| 2026-06-03 | Claude Code Plugin Marketplace Landscape | Preparing to publish llm-tutor to a marketplace, needed to know what marketplaces actually exist | llm-tutor release plan (submission order: topic tag → community marketplace → hesreallyhim PR → Composio PR), README install instructions, GitHub repo Topics field |
The 2026-06-02 catalog entry above is the canonical record. These five docs are deeper per-angle bibliographies referenced by the catalog:
| File | Angle |
|---|---|
01-how-boots-actually-works.md |
Boots internal mechanics, prompting model, failure modes, economic data |
02-llm-tutor-landscape-2026.md |
Commercial + research LLM tutors (Khanmigo, Study Mode, TutorAI, academic patterns) |
03-pedagogy-and-curriculum-theory.md |
Bloom's, ZPD, scaffolding, Knowledge Space Theory, Productive Failure |
04-gamification-evidence.md |
Bjork, Ryan & Deci (SDT), Duolingo critiques, salmon-cost justification |
05-anti-dependency-design.md |
Illich, MIT EEG, convivial tools, bicycle vs e-bike framing |
When you research something external (vendor docs, papers, blog posts, open-source repos), the /flagrare:research-catalog skill produces a file in this directory and a row in this table. Run it before the synthesis goes back to the requester.