Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Graphify

ASecurity

Use when answering questions about a large/legacy codebase burns tokens or hallucinates, or when onboarding to a new repo. Triggers on "graphify", "grafo de conhecimento", "mapa do repositório", "entender a arquitetura", "consome muitos tokens", "como X conecta com Y", "onboarding em projeto". Turns a repo into a queryable knowledge graph.

2 stars
0 votes
0 copies
0 views
Added 9/19/2026
ai-agentsgonodegit

Security Analysis

A100/100

Scanned 9/19/2026

Install to Claude Code

$npx -y skills add majinmagros/magros.ai-skills --skill graphify --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Graphify?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Graphify
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/majinmagros-graphify/badge)](https://www.skillsdirectory.com/skills/majinmagros-graphify)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: graphify
description: Use when answering questions about a large/legacy codebase burns tokens or hallucinates, or when onboarding to a new repo. Triggers on "graphify", "grafo de conhecimento", "mapa do repositório", "entender a arquitetura", "consome muitos tokens", "como X conecta com Y", "onboarding em projeto". Turns a repo into a queryable knowledge graph.
---

# Skill: Graphify — repositório vira grafo de conhecimento

Transforma um repositório inteiro (código, docs, PDFs, áudio, vídeo) num
grafo de conhecimento consultável. Em vez de varrer arquivo por arquivo
(grep/ctrl+F cegos queimando tokens e errando), o agente consulta o mapa.

## Quando Ativar (gatilhos literais)

- "Graphify esse repo"
- "Como X conecta com Y no código?"
- "Mapa do repositório antes de mexer"
- "Onboarding nesse projeto herdado"
- "Resposta queimando token demais"

## Exemplo

```text
/graphify explain "como o login chega na tabela users?"
→ form → POST /login → authService → users(id)
```

## 1. Quando usar

- Perguntas sobre projeto grande: "como o login conecta com a tabela de
  usuários?" (cadeia form → endpoint → serviço → tabela).
- Onboarding em projeto novo ou herdado.
- Respostas com alucinação porque o agente não viu as conexões reais.
- Economizar tokens em consultas frequentes sobre a mesma base.

## 2. RAG vetorial vs grafo (tradeoff)

| | RAG vetorial | Grafo |
|---|---|---|
| Responde | "o que é parecido" | "como as coisas se conectam" |
| Ideal | Textos soltos por tema | Arquitetura/dependências |
| Mudança | Re-embeda TUDO de novo | Grafo vivo: atualiza só o que mudou |

## 3. Instalação/uso

- Open source, MIT, ~70k★ (`graphify` no GitHub). Funciona com qualquer agente.
- Vem com **skill embutida** que ensina o próprio Claude a usar.
- Comandos principais: `/graphify` (indexa o diretório), `/graphify explain`
  (consulta o grafo), `/graphify claude install` (fica sempre ligado, consulta
  o grafo em toda pergunta), flag Obsidian (`/graphify obsidian`).
- 3 camadas de indexação: estrutura de código (determinística, sem LLM) →
  áudio/vídeo (whisper) → docs/imagens (semântica com LLM).

## 3b. Alternativa GRAFT (Batch 17f, #72)

GRAFT (open source, MIT): code graph via tree-sitter (structural layer,
sem AI key, sem DB server, `.graft/` gitignored) + optional AI deep
build (summaries + concept nodes, custa tokens). Mesmo trabalho do
graphify para perguntas de estrutura; escolha um grafo por repo.
Fonte: video `VW58Q8c5F0I` (demo via Verdant).

## 4. Processo

1. Confirme que o repo está indexado (ou rode `/graphify`).
2. Para perguntas de conexão/arquitetura: force consulta ao grafo
   ("olhe o grafo antes de responder").
3. Entregue a resposta com o caminho das conexões (nó → aresta → nó).
4. Se o código mudou muito, re-indexe (grafo vivo atualiza só o afetado).

## 5. Regra

- Grafo é para RELAÇÕES. Conteúdo solto por tema ainda é caso de vetorial —
  não force grafo onde RAG resolve.

Attribution

majinmagrosmajinmagros
View sourceMore from majinmagros →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

1023331 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3331 votes

catchup

Recovers prior coding-agent session context by running `catchup <agent> --since-compact`, which extracts a clean summary of a previous Codex, Claude Code, Antigravity, OpenCode, or Pi Agent session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", or asks to recover/summarize a previous session before continuing. Do NOT use for the current conversation, git history, or any non-agent log.

611 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →