convez.ai
SkillAutomaçãoIntermediário
Bruno-Cunha-Souza

ValarMindSkills

Bruno-Cunha-Souza/ValarMindSkills

Biblioteca de skills para agentes IA — desenvolvimento, segurança, otimização

Ver no GitHub
BomPluginScore · de 0 a 100

Biblioteca de skills reutilizáveis para agentes de IA, estruturada como arquivos Markdown com frontmatter YAML. Integra-se nativamente com Claude Code CLI, Cursor IDE e outras ferramentas, permitindo invocar 17+ skills especializadas via comando slash para desenvolvimento, automação e análise de código.

Instalação

$npx convez-ai Bruno-Cunha-Souza/ValarMindSkills

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
5
Forks
0
Acompanhando
0
Saúde
42%
Último commit
há 1 mês
Versão
v3.1.0
Issues abertas
0
Contribuidores
1

Casos de uso

  • Geração automática de CI/CD (GitHub Actions) com detecção de linguagem
  • Análise de segurança em ciclo de vida para web, APIs, Go, Next.js e Python
  • Otimização de performance e revisão de código com severity ranking
  • Geração de release notes e commit messages (Conventional Commits)
  • Debugging especializado e aplicação de Clean Code principles

Veredito

ValarMindSkills é uma solução madura e bem-documentada para times que usam Claude Code CLI ou Cursor IDE. Cobre fluxos reais de desenvolvimento (CI/CD, segurança, otimização, code review) com footprint de token transparente (~6.5k combinado). Recomendado para equipes que querem estender capabilities de IA com workflows profissionais.

README

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

ValarMindSkills

A library of reusable skills for AI agents. Each skill is a Markdown file with YAML frontmatter that can be invoked as a slash command within Claude Code CLI, Cursor IDE, Codex CLI, or Antigravity IDE.

Quick install (recommended)

Run the unified installer to set up skills for Claude Code CLI, Codex CLI, Antigravity, and Cursor IDE. The script is idempotent, so you can re-run it anytime to upgrade.

curl -fsSL https://raw.githubusercontent.com/Bruno-Cunha-Souza/ValarMindSkills/main/install.sh | bash

Token usage

The following table estimates token consumption per skill/context at session start. Values are approximate.

ComponentTokens
Skill descriptions (17 skills, YAML frontmatter)~ 970
Caveman (SessionStart hook)~ 2,000
Superpowers (SessionStart hook)~ 3,400
Obsidian-brain (SessionStart hook)~ 160
Per-turn reinforcement (UserPromptSubmit)~ 90

Scenarios

ScenarioTokens
Skill descriptions only~ 970
Descriptions + Caveman active~ 2,970
Descriptions + Superpowers active~ 4,400
Descriptions + Obsidian-brain active~ 1,130
All combined~ 6,530

Context impact: 6,530 tokens ≈ 0.65% of a 1M context window or 2.49% of a 262k window.


Available skills

SkillDescription
cavemanTerse response mode — drops articles, filler, hedging. Intensity levels: lite / full / ultra
ci-cd-generatorGitHub Actions CI/CD generator for Go/Rust/TS — auto-detects language, encodes coverage/race/leak gates, wires SAST/SCA/secret/container/SBOM scans by security level
clean-codeApplies Clean Code principles for quality, readability, and maintainability
code-debuggerDebugging specialist for errors, test failures, and unexpected behavior
code-optimizationLifecycle performance & efficiency audit for Go (Gin/Fiber/fx), Rust (Axum/Actix/Tokio), TypeScript (Node/Bun/Elysia/Fastify), Python (FastAPI/Django/Flask, CPython 3.13/3.14 incl. free-threaded 3.14t) — profiling, duplication/leaks/N+1/CPU-bound/blocking-I/O sweeps, tri-axis Impact × Risk × Effort classification, optional WebSearch + context7 validation, writes OPTIMIZATION_REPORT.md at project root
code-reviewLifecycle code review (Go/Rust/TS) with severity-ranked findings, risk tags, and references/NEXTJS.md for Next.js 16.2.x performance audits (RSC, <img>, Turbopack)
context-optimizationApply compaction, masking, and caching strategies
github-release-noteGenerates release notes from a git tag range
github-commitGenerates commit messages following Conventional Commits
github-pr-reviewPerforms structured code review of Pull Requests
code-security-reviewWeb+API+Go+Next.js+Python security lifecycle (FastAPI, Gin, Fiber, Elysia, Next.js 16 App Router) — Phase 0 stack + AI/CI-surface detection branches into references/golang/ or references/nextjs/; generic design controls + active testing + 100-vuln catalog + supply-chain/CI-CD hardening + LLM/agentic/MCP coverage. OWASP Web Top 10 2025 + API 2023 + LLM 2025 + Agentic 2026; static probes for deps, CI workflows, secrets, AI
obsidian-basesCreate and edit Obsidian Bases (.base files) with views, filters, formulas, and summaries
obsidian-brainToken-efficient session-memory for Obsidian-vault projects — auto-detects, lazy-loads index, writes atomic notes for sessions/topics/decisions. ON by default when vault detected
obsidian-cliInteract with Obsidian vaults via the Obsidian CLI — read, create, search, manage notes and plugins
obsidian-markdownObsidian Flavored Markdown reference — wikilinks, embeds, callouts, properties
only-planRead-only planning modifier — never edits project files; writes a step-by-step implementation plan to a single new IMPLEMENTATION_PLAN.md at the project root. Composable with other skills: /code-security-review /only-plan
prompt-engineeringAudits, hardens, and rewrites LLM prompts (SKILL.md, RAG, tool descriptions, agent base prompts) — severity-ranked findings with risk tags, rewritten prompt, token delta
skill-creatorMeta-skill that scaffolds new skills for this repository following project conventions
superpowersEngineering-discipline posture — 1% skill-scan, four pillars (TDD, systematic, complexity, evidence), seven-stage workflow. OFF by default

Installation on Claude Code CLI

Plugin install (manual, from source)

Registers the repository as a local Claude Code marketplace and installs valarmindskills@valarmindskills. Brings all 19 skills under the /valarmindskills:<slug> namespace and enables the caveman auto-activation hooks (SessionStart + UserPromptSubmit), plus the obsidian-brain hooks (SessionStart detection + UserPromptSubmit toggle, ON by default).

git clone https://github.com/Bruno-Cunha-Souza/ValarMindSkills.git
cd ValarMindSkills
bash scripts/install-plugin-claude.sh

To install in Claude Code CLI, Codex CLI, Antigravity, and Cursor IDE in one command, use the unified installer:

bash scripts/install-all.sh

To uninstall: claude plugins uninstall valarmindskills@valarmindskills && claude plugins marketplace remove valarmindskills.

Alternative: development load (no install)

# Reads the repo in place. Run /reload-plugins after edits.
claude --plugin-dir /path/to/ValarMindSkills

After install, open a new session and caveman mode activates at level lite by default. Control with:

  • /valarmindskills:caveman lite|full|ultra — switch intensity
  • /valarmindskills:caveman off — deactivate
  • stop caveman / normal mode (natural language) — deactivate

Override the default mode with CAVEMAN_DEFAULT_MODE=lite in your environment, or with defaultMode in ~/.config/caveman/config.json.

Superpowers (off by default)

The plugin also ships a superpowers posture inspired by obra/superpowers: scan skills before each reply (1% rule), follow the user > skills > defaults hierarchy, refuse twelve rationalizations, apply four pillars (TDD, systematic, complexity reduction, evidence), and walk a seven-stage workflow when scope warrants it. Unlike caveman, superpowers is OFF by default; you opt in.

  • /valarmindskills:superpowers on (or bare /valarmindskills:superpowers) — activate for the current session
  • /valarmindskills:superpowers off — deactivate
  • stop superpowers / desativar superpowers (natural language) — deactivate

Make activation persistent with SUPERPOWERS_DEFAULT_MODE=on in your environment, or with {"defaultMode": "on"} in ~/.config/superpowers/config.json.

Caveman and superpowers coexist freely — caveman shapes voice, superpowers shapes process. When both are active, the statusline renders both badges ([CAVEMAN] | [SUPERPOWERS] | context: …).

Obsidian Brain (on by default when vault detected)

Token-efficient session-memory for any project whose CLAUDE.md or AGENTS.md references an Obsidian vault. The SessionStart hook auto-detects the vault, writes a flag file at ~/.claude/.obsidian-brain-active, and injects a one-time digest pointing the agent at the brain index. With no vault detected, the hook silently clears the flag and the statusline badge hides — no posture, no noise.

  • /valarmindskills:obsidian-brain on (or bare /valarmindskills:obsidian-brain) — re-enable for the current session
  • /valarmindskills:obsidian-brain off — disable for the current session
  • stop obsidian-brain / desativar obsidian-brain (natural language) — deactivate

Disable persistently via OBSIDIAN_BRAIN_DEFAULT_MODE=off in your environment, or with {"defaultMode": "off"} in ~/.config/obsidian-brain/config.json.

Statusline

The plugin ships a composable statusline that combines three optional badges with the current context window usage (e.g. 42% 420k/1M, color-coded by threshold):

  • [CAVEMAN] / [CAVEMAN:ULTRA] — laranja, hidden when caveman is off.
  • [SUPERPOWERS] — cyan when on, dim when off (always visible).
  • [OBSIDIAN-BRAIN] — roxo (cor 99 ≈ #875FFF, próxima do roxo Obsidian), hidden when no vault is detected or the user opted out.

scripts/install-plugin-claude.sh configures it automatically: it adds statusLine to ~/.claude/settings.json (creating the file if needed, backing it up if it already exists). If statusLine is already set to a different command, the installer leaves it untouched and prints both values so you can choose. Set VALARMIND_SKIP_STATUSLINE=1 to opt out, or remove it manually:

"statusLine": {
  "type": "command",
  "command": "bash \"/path/to/ValarMindSkills/hooks/statusline/statusline.sh\""
}

The statusline is built from independent segments under hooks/statusline/segments/ — additional segments can be added there without touching the Caveman plugin.


Installation on Antigravity IDE

Note: Antigravity does not load symlinks. Skills must be copied as real files.

Install script for Antigravity (recommended)

The repository includes a script that copies the skills to the global Antigravity directory:

git clone https://github.com/Bruno-Cunha-Souza/ValarMindSkills.git
cd ValarMindSkills
bash scripts/install-antigravity.sh

To install in Claude Code CLI, Codex CLI, and Antigravity in one command, use the unified installer:

bash scripts/install-all.sh

After running the script, reload the VS Code window so the autocomplete picks up the new skills:

  • Cmd + Shift + P > Reload Window

Re-run bash scripts/install-antigravity.sh (or bash scripts/install-all.sh) whenever you pull new changes or add/edit skills.

Per-project installation (optional)

To limit skills to a specific project, copy them into the project root:

mkdir -p .agent/skills
cp -r ValarMindSkills/skills/* .agent/skills/

Installation on Codex CLI

Plugin install (recommended)

Copies all skills to ~/.codex/skills/, copies the caveman / superpowers / obsidian-brain hook scripts to ~/.codex/hooks/, and injects the corresponding [[hooks.SessionStart]] and [[hooks.UserPromptSubmit]] entries into ~/.codex/config.toml. Also writes the matching postures into ~/.codex/AGENTS.md. Both the config.toml block and the AGENTS.md block are wrapped in # >>> VALARMIND BEGIN/END (or <!-- VALARMIND BEGIN/END -->) markers, so re-running the installer rewrites the managed block in place without duplicating entries.

git clone https://github.com/Bruno-Cunha-Souza/ValarMindSkills.git
cd ValarMindSkills
bash scripts/install-plugin-codex.sh

Override the target home with CODEX_HOME=/custom/path bash scripts/install-plugin-codex.sh.

Skills-only install (lite)

For a setup without hooks or AGENTS.md postures (just the slash-command surface):

bash scripts/install-codex.sh

Restart Codex CLI after either script so the new skills and hooks are picked up.


Installation on Cursor IDE

Plugin install (recommended)

Copies all skills to ~/.cursor/skills/, copies hook scripts to ~/.cursor/hooks/, and merges ValarMind entries into ~/.cursor/hooks.json (sessionStart + beforeSubmitPrompt). Session-start hooks run through hooks/_cursor/wrap-session.sh, which adapts Claude-style plain-text output to Cursor's JSON additional_context format. Re-running the installer removes prior ValarMind hook entries before appending fresh ones (no duplicates).

git clone https://github.com/Bruno-Cunha-Souza/ValarMindSkills.git
cd ValarMindSkills
bash scripts/install-plugin-cursor.sh

Override the target directory with CURSOR_HOME=/custom/path bash scripts/install-plugin-cursor.sh.

Set VALARMIND_SKIP_HOOKS=1 to install skills only (no hooks.json changes).

Skills-only install (lite)

bash scripts/install-cursor.sh

Restart Cursor after either script. Check Settings → Hooks (or the Hooks output channel) if hooks do not load.

Modes in Cursor

Skills are invoked with @slug (e.g. @code-review, @caveman). Caveman, superpowers, and obsidian-brain postures behave like Codex/Claude:

  • Caveman — ON by default at level lite. Toggle with natural language (stop caveman, normal mode) or by mentioning @caveman.
  • Superpowers — OFF by default. Activate with @superpowers or phrases like superpowers on.
  • Obsidian-brain — ON when the workspace CLAUDE.md or AGENTS.md references an Obsidian vault path.

Override defaults with CAVEMAN_DEFAULT_MODE, SUPERPOWERS_DEFAULT_MODE, or OBSIDIAN_BRAIN_DEFAULT_MODE, or the JSON files under ~/.config/caveman/, ~/.config/superpowers/, and ~/.config/obsidian-brain/.

Cursor does not support the Claude Code statusline; posture state is stored in flag files under ~/.cursor/ (e.g. ~/.cursor/.caveman-active).

Third-party Claude hooks (optional)

If you already use Claude Code hooks in ~/.claude/settings.json, you can enable Settings → Features → Third-party skills to load them in Cursor in parallel. The ValarMind Cursor installer uses native ~/.cursor/hooks.json and does not require this setting.


Project structure

.claude-plugin/
  plugin.json               <- plugin manifest (name, hooks)
  marketplace.json          <- local marketplace manifest
hooks/
  caveman/
    caveman-activate.js             <- SessionStart hook (on by default)
    caveman-mode-tracker.js         <- UserPromptSubmit hook
    caveman-config.js               <- shared helpers
  superpowers/
    superpowers-activate.js         <- SessionStart hook (off by default)
    superpowers-mode-tracker.js     <- UserPromptSubmit hook
    superpowers-config.js           <- shared helpers
  obsidian-brain/
    obsidian-brain-activate.js      <- SessionStart hook (on when vault detected)
    obsidian-brain-mode-tracker.js  <- UserPromptSubmit hook
    obsidian-brain-config.js        <- shared helpers
  _cursor/
    wrap-session.sh                 <- Cursor sessionStart JSON adapter
  _lib/
    resolve-skill-path.js           <- shared SKILL.md path resolver
  statusline/
    statusline.sh                   <- composer (entry registered in settings.json)
    segments/
      caveman.sh                    <- caveman mode badge segment
      superpowers.sh                <- superpowers mode badge segment
      obsidian-brain.sh             <- obsidian-brain badge segment (roxo)
      context.sh                    <- context window usage segment
skills/
  <slug>/
    SKILL.md                <- skill definition (YAML frontmatter + Markdown instructions)
scripts/
  install-plugin-claude.sh  <- persistent plugin install for Claude Code CLI
  install-plugin-codex.sh   <- full plugin install for Codex CLI (skills + hooks + AGENTS.md)
  install-codex.sh          <- skills-only install for Codex CLI (no hooks)
  install-plugin-cursor.sh  <- full plugin install for Cursor IDE (skills + hooks.json)
  install-cursor.sh         <- skills-only install for Cursor IDE (no hooks)
  install-antigravity.sh    <- copies skills to Antigravity global directory
  install-all.sh            <- runs all plugin installers (Claude + Codex + Antigravity + Cursor)
install.sh                  <- curl-bash bootstrap (downloads latest release, runs install-all.sh)

Each directory under skills/ represents a skill. The directory slug is the identifier used as a slash command.


Skill format

Each SKILL.md follows this format:

---
name: skill-name
description: Short description of when the skill should be triggered.
source: ValarMind Skills
---

Followed by Markdown sections:

  • Goal — what the skill does
  • Inputs — data the agent must collect before executing
  • Procedure — step-by-step execution instructions
  • Constraints — limits and rules
  • Output format — how results should be presented

Contributing

  1. Create a directory under skills/ with the new skill's slug
  2. Add a SKILL.md file following the format above
  3. Write instructions in English
  4. Open a pull request

License

MIT

Ver o original no GitHub

Bom para

Relacionados