brave-search/SKILL.md
Web search via Brave Search with hybrid mode - API primary with web scraping fallback. Use for searching documentation, facts, or any web content. Automatically switches to scraping if API unavailable.
npx skillsauth add alweil/openclaw-pessoal brave-searchInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Web search via Brave Search API with automatic fallback to web scraping. No browser required.
| Method | Priority | When Used | |--------|----------|-----------| | API | Primary | Quando a API key é válida e tem orçamento | | Web Scraping | Fallback | Quando API retorna erro (403, 429, limite excedido) |
| Script | Purpose |
|--------|---------|
| brave.js | Híbrido (recomendado) - Tenta API primeiro, fallback para scrape |
| search.js | Web scraping puro (original) |
| brave-api.js | API puro |
| content.js | Extrair conteúdo de URLs específicas |
# Híbrido - automático
./brave.js "javascript async await"
# Com opções
./brave.js "restaurantes sp" -n 10 -c BR -l pt
./brave.js "query" # Basic search (5 results)
./brave.js "query" -n 10 # More results (max 20)
./brave.js "query" -c BR # Country code (BR, US, etc)
./brave.js "query" -l pt # Language (pt, en, etc)
📊 Results for: "javascript async await" (5 found via 🔌 API)
--- Result 1 [API] ---
Title: JavaScript Async/Await
Link: https://developer.mozilla.org/...
Snippet: The async function declaration...
--- Result 2 [API] ---
...
Ou se a API falhar:
⚠️ API failed (403: Forbidden), trying web scraping...
📊 Results for: "query" (5 found via 🌐 Scrape)
Se quiser garantir uso da API (falha se indisponível):
./brave-api.js "query" -n 10 -c US -l en
Se quiser garantir uso do scraping:
./search.js "query" # Basic
./search.js "query" -n 10 # More results
./search.js "query" --content # Include page content
Extrai conteúdo legível de uma URL:
./content.js https://example.com/article
Instalar dependências (necessário para web scraping):
npm ci
A API key está configurada no script. Para usar sua própria key:
export BRAVE_API_KEY="sua-key-aqui"
./brave-api.js "query"
Ou edite a linha no topo de brave-api.js ou brave.js.
| Feature | API | Scrape | |---------|-----|--------| | Velocidade | ⚡ Mais rápido | 🐢 Mais lento | | Confiabilidade | ✅ Estrutura garantida | ⚠️ Pode quebrar se site mudar | | Rate limit | 2000/mês grátis | Ilimitado (mas lento) | | Custo | Grátis até limite | Sempre grátis | | Precisão | ✅ Alta | ✅ Alta |
content-media
Fetch and summarize YouTube video transcripts. Use when asked to summarize, transcribe, or extract content from YouTube videos.
tools
Work with Obsidian vaults (plain Markdown notes) and automate via obsidian-cli.
content-media
Generate/edit images with Nano Banana Pro (Gemini 3 Pro Image). Use for image create/modify requests incl. edits. Supports text-to-image + image-to-image; 1K/2K/4K; use --input-image.
tools
Centralized Activity Feed for all OpenClaw agents. Captures timestamped actions including tool calls, file writes/edits, memory updates, task completions, agent identification, and mental models. Use when tracking agent activities across sessions, auditing operations, building activity timelines, or implementing observability for multi-agent workflows. Works for all agents (main, trabalho, projetos, pessoal, etc.).