convez.ai
SkillBusca & PesquisaIntermediário
academicatstool-netizen

Find Angles — Mapeador de Direções de Pesquisa

academicatstool-netizen/Cat_find_angles

Mapeie direções de pesquisa distintas com sub-ângulos e queries prontas

Ver no GitHub
InicialPluginScore · de 0 a 100

Transforma um tema vago em uma árvore estruturada de direções de pesquisa distintas, cada uma com sub-ângulos, palavras-chave e queries prontas para executar. Mantém o foco ancorado às suas próprias palavras, sem desvios temáticos, e funciona em Claude, ChatGPT e DeepSeek.

Instalação

$npx convez-ai academicatstool-netizen/Cat_find_angles

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
1
Forks
0
Acompanhando
0
Saúde
42%
Último commit
há 5 meses
Versão
—
Issues abertas
0
Contribuidores
2

Casos de uso

  • Explorar múltiplas vertentes de um tema de pesquisa vagamente definido
  • Gerar queries de busca académica estruturadas por direção e sub-ângulo
  • Evitar desvios temáticos ao expandir um tópico em linhas concretas de investigação

Veredito

Skill sólida e bem-pensada para pesquisadores que precisam estruturar tópicos amplos em direções investigáveis. O rigor em manter ancoragem às palavras originais e a entrega de queries prontas a diferencia de abordagens genéricas de brainstorm. Implementação em shell é funcional, mas o produto completo no AcademiCats será a referência.

README

A tradução para o português deste README ainda está na fila.

English · 中文

🐱 Find Angles

Turn one vague topic into a map of the field — distinct research directions, concrete sub-angles, and a search query for each. For when you have a topic but don't know where to focus.

License: MIT  Claude  ChatGPT  DeepSeek  Full product


🪶 A free, open-source preview of a capability coming to AcademiCats

Note: unlike the other skills in this family, direction-finding isn't in the current AcademiCats product yet — it's planned for a future official release. This open skill is an early, self-contained version of that first step — turning a topic into research directions — that you can already run today on your own AI: Claude, ChatGPT, or DeepSeek.


✨ What it does

🧭 Breadth without drift — it maps the genuinely distinct lines of work on your topic (4–8 of them), while keeping every direction anchored to your own words. No wandering from "AR games" into "gamification in healthcare" — the failure mode it's built to prevent.

🌳 A tree, not a list — each big direction expands into 4–7 concrete sub-angles, so you can see both the forest and the specific studies you could actually run.

🔍 Every node ships a search query — each direction and sub-angle comes with a real academic-database query you can paste straight into Paper Search (or Google Scholar / PubMed) to pull the papers for that angle.

🚫 No made-up findings — it proposes directions and the queries to investigate them; it never claims a paper or result exists. Pure reasoning — runs fully offline, on any model.

🎬 Demo

"find angles for XR experience"

It runs a three-stage pipeline internally (decompose → expand → build) and returns a fixed-shape directions tree — the same shape no matter which model runs it:

🧭 Research directions for: XR experience

Core terms: XR experience · extended reality · user experience · Intent: exploratory

1. Measuring XR user experience ⭐

How to validly capture user experience in extended reality — presence, comfort, usability. Keywords: extended reality · user experience · presence · questionnaire 🔍 extended reality user experience measurement presence

  • 1a · Presence & immersion measures — isolates the XR-specific driver of experience · 🔍 extended reality presence immersion questionnaire
  • 1b · Cybersickness & comfort — the dominant UX failure mode in XR · 🔍 extended reality cybersickness comfort user experience
  • 1c · Standardized XR UX questionnaires — which scales validly capture it · 🔍 extended reality user experience questionnaire validation

2. XR experience in museums & cultural heritage

…


🧭 6 directions · 26 sub-angles · recommended ⭐ #1 → "Measuring XR user experience" · paste any 🔍 query into Paper Search or a database

Every title carries one of your core terms; nothing drifts into an unrelated field; and "paste 1b into Paper Search" is all it takes to start pulling real papers for that angle.

🚀 Get started — pick your platform

Pick whichever AI you use; each setup takes under a minute.

🖥️ Claude Code — runs locally, triggers itself

mkdir -p ~/.claude/skills
git clone https://github.com/academicatstool-netizen/Cat_find_angles.git ~/.claude/skills/find-angles

Restart Claude Code, then just ask — "help me find research directions for …"

🌐 Claude (web / desktop app) — download find-angles.skill, then upload it under Settings → Capabilities → Skills. Ask in any chat.

🤖 ChatGPT — open PORTABLE_PROMPT.md and paste it into a Custom GPT's Instructions (or just send it as your first message). Then send your topic.

💬 DeepSeek / any other model — paste PORTABLE_PROMPT.md as the system prompt (or first message). Then send your topic.

✅ No web access needed — Find Angles is pure reasoning, so it runs fully on any model, online or off. You bring the topic; it never needs to look anything up.

💙 What's inside

A three-stage pipeline — decompose → expand → build directions — a preview of the "Query Understanding" engine we're building into the full product, run as sequential reasoning passes in one turn.

Find Angles (this skill)AcademiCats (planned) →
⚡ Speedseconds (one model pass)seconds — streamed, per-direction
Anchored, drift-free directions✅✅
Ready-to-run search query per node✅✅
One-click "search this direction"paste into Paper Search✅ end-to-end
Saved direction history—✅
Polished web & mobile app—✅

🐱 The AcademiCats skill family

Four open skills that chain into one research workflow — install any or all:

  • 🧭 Find Angles (you are here) — turn a topic into research directions
  • 🔍 Paper Search — find & read papers
  • ✍️ Synthesis Lab — write grounded papers from your sources
  • 🧪 Paper Review — peer-review your own draft

Install all at once — clone any one repo, then run bash install.sh.

🙋 FAQ

  • What are all these files? Use just one path above — a git clone (Claude Code), the .skill file (Claude web/desktop), or PORTABLE_PROMPT.md (ChatGPT/DeepSeek). SKILL.md and references/ are internals your assistant loads for you — no need to open them.
  • It didn't trigger? Restart Claude Code after installing, and phrase your message as a task — "find research directions for …", "what angles can I take on …".
  • Does it find papers? No — it maps directions and hands you a search query for each. Pair it with Paper Search to pull the real papers for any direction.
  • Which model? Any strong model works — Claude Sonnet/Opus, GPT‑4o/o‑series, or DeepSeek‑V3/R1 give the best results.
  • Private & free? It runs on your own AI — no account, nothing sent to us, no web access required.

Want the whole research workflow?

→ academicats.com ←

🚀 The full product is in open beta — free to try right now.

Made with 💙 by the AcademiCats team · MIT License

Ver o original no GitHub

Bom para

Relacionados