Search First
shimo4228/search-first
Pesquise soluções antes de reinventar a roda
Ver no GitHubUma skill para agentes que força a pesquisa de soluções existentes antes de codificar. Busca em npm, PyPI, MCP servers e GitHub para avaliar se deve adotar uma biblioteca pronta, estender uma existente, compor múltiplos pacotes ou construir uma solução customizada.
Instalação
npx convez-ai shimo4228/search-first1 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
Casos de uso
- Iniciar novo recurso sem reimplementar código existente
- Avaliar e selecionar dependências com base em critérios técnicos
- Reduzir débito técnico e fragmentação de código
Veredito

Skill bem fundamentada que automatiza uma disciplina crítica — pesquisa antes de desenvolvimento. Implementa metodologia sólida (Agent Knowledge Cycle) com matriz de decisão clara (adotar/estender/compor/construir), reduzindo reinvenção real em fluxos de agentes.
README
A tradução para o português deste README ainda está na fila.
Language: English | 日本語
search-first
An Agent Skill that enforces a research-before-coding workflow. Before writing custom code, the agent searches for existing tools, libraries, MCP servers, and patterns — then makes an informed adopt/extend/build decision.
Install
Claude Code
# Copy into your global skills directory
cp -r skills/search-first ~/.claude/skills/search-first
SkillsMP
/skills add shimo4228/search-first
How It Works
- Need Analysis — Define what functionality is needed
- Parallel Search — Search npm/PyPI, MCP servers, GitHub, and existing skills
- Evaluate — Score candidates on functionality, maintenance, community, docs, license, and deps
- Decide — Adopt as-is, extend/wrap, compose multiple packages, or build custom
- Implement — Install the chosen solution with minimal custom code
When It Triggers
- Starting a new feature that likely has existing solutions
- Adding a dependency or integration
- Before creating a new utility, helper, or abstraction
Decision Matrix
| Signal | Action |
|---|---|
| Exact match, well-maintained, MIT/Apache | Adopt — install and use directly |
| Partial match, good foundation | Extend — install + write thin wrapper |
| Multiple weak matches | Compose — combine 2-3 small packages |
| Nothing suitable found | Build — write custom, but informed by research |
Examples
Need: Check markdown files for broken links
Search: npm "markdown dead link checker"
Found: textlint-rule-no-dead-link (score: 9/10)
Action: ADOPT — npm install textlint-rule-no-dead-link
Result: Zero custom code, battle-tested solution
About this skill
This skill implements the Research phase of the Agent Knowledge Cycle (AKC) — a Zenodo-citable six-phase bidirectional growth loop (DOI 10.5281/zenodo.19200726) for sustaining intent alignment between an AI agent and its operator over time. AKC is one of three research lines by @shimo4228, alongside Contemplative Agent (DOI 10.5281/zenodo.19212118) — autonomous agents grounded in four contemplative axioms — and Agent Attribution Practice (AAP) (DOI 10.5281/zenodo.19652013) — harness-neutral ADRs on accountability distribution.
License
MIT