npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@evolve.labs/devflow

v1.3.1

Published

Multi-agent system for software development with Claude Code. 6 specialized agents (Strategist, Architect, System Designer, Builder, Guardian, Chronicler) as slash commands.

Readme

DevFlow v1.3.1 - Sistema Multi-Agentes para Desenvolvimento de Software

Sistema de 7 agentes especializados para desenvolvimento de software com Claude Code, com Autopilot para execução automatizada do pipeline completo.

Version npm License: MIT


Instalacao

Via npm (Recomendado)

# Instalar globalmente
npm install -g @evolve.labs/devflow

# Inicializar no seu projeto
devflow init /caminho/para/seu-projeto

# Ou usar npx (sem instalar)
npx @evolve.labs/devflow init

Opcoes de instalacao

devflow init                    # Agentes + estrutura de docs (padrao)
devflow init --agents-only      # Apenas agentes (minimo)
devflow init --full             # Tudo incluindo .gitignore

Comandos

devflow init [path]             # Inicializar DevFlow num projeto
devflow update [path]           # Atualizar instalacao existente
devflow autopilot <spec-file>   # Rodar autopilot nos agentes
devflow challenge [review-file] # Review adversarial com OpenAI (standalone)

Flags do devflow update:

devflow update                          # Atualiza se houver nova versao
devflow update --force                  # Re-copia todos os arquivos mesmo se versao ja e igual
devflow update --source <path>          # Usa repo local como fonte (dev workflow)
devflow update --force --source <path>  # Forcado + repo local

Dev workflow (apos editar o repo local): npm run reinstall para atualizar o CLI global, depois devflow update no projeto.

Requisitos

  • Claude Code CLI (npm i -g @anthropic-ai/claude-code)
  • Node.js 18+
  • Git (recomendado)

Os 7 Agentes

| # | Agente | Funcao | Uso | | --- | --------------------------- | --------------------------- | ---------------------------------------- | | 1 | /agents:strategist | Planejamento & Produto | Requisitos, PRDs, user stories | | 2 | /agents:architect | Design & Arquitetura | Decisoes tecnicas, ADRs, APIs | | 3 | /agents:system-designer | System Design & Escala | SDDs, RFCs, capacity planning, SLOs | | 4 | /agents:builder | Implementacao | Codigo, reviews, refactoring | | 5 | /agents:guardian | Qualidade & Seguranca | Testes, security, performance | | 6 | /agents:challenger | Review Adversarial (OpenAI) | Desafia o Guardian, encontra blind spots | | 7 | /agents:chronicler | Documentacao & Memoria | CHANGELOG, snapshots, stories |

Cada agente tem hard stops — limites rigidos que impedem de fazer trabalho de outros agentes.

Challenger requer OPENAI_API_KEY — usa OpenAI o3 como perspectiva independente do Guardian.

Fluxo de Trabalho

strategist → architect → system-designer → builder → guardian → challenger → chronicler

Dois Modos de Scaling por Agente

Modo Padrao — Parallel Subagents

Ativado automaticamente (sem argumento extra). Usa o Agent tool do Claude para criar workers isolados em paralelo.

/agents:builder implemente autenticacao JWT

Modo Team — Claude Agent Teams

Ativado com o argumento team. Usa Claude Agent Teams (peers que se comunicam diretamente entre si).

/agents:builder team implemente autenticacao JWT

devflow init configura isso automaticamente. Para instalaçoes existentes: devflow update.

| | Modo Padrao | Modo Team | | ----------- | ------------------------- | -------------------------- | | Trigger | sem argumento | argumento team | | Comunicacao | Pai → Filho | Peers direto | | Setup | Automatico | Flag experimental | | Custo | 1x tokens | 3-5x tokens | | Quando usar | Sub-tarefas independentes | Debate/revisao entre peers |

Use /agents:team para orquestracao cross-domain com todos os 6 agentes em paralelo.


Autopilot

O autopilot executa os agentes em sequencia automaticamente com Persistent Memory, Adaptive Planning e Context Isolation.

# Rodar todas as fases
devflow autopilot docs/specs/minha-spec.md

# Adaptive: LLM decide quais fases rodar (TRIVIAL/SIMPLE/MODERATE/COMPLEX)
devflow autopilot docs/specs/minha-spec.md --adaptive

# Com Challenger (review adversarial OpenAI apos guardian)
devflow autopilot docs/specs/minha-spec.md --challenger

# Fases especificas
devflow autopilot docs/specs/minha-spec.md --phases "strategist,architect,builder"

# Desativar context isolation (usa todo o historico)
devflow autopilot docs/specs/minha-spec.md --full-context

Flags do autopilot

--phases <list>             # Fases especificas (default: todos os 6 agentes)
--adaptive                  # LLM escolhe fases por complexidade
--full-context              # Desativa context isolation (N-1 output por padrao)
--challenger                # Review OpenAI apos guardian (requer OPENAI_API_KEY)
--challenger-model <model>  # gpt-5.4 (default), o3-mini, o3
--no-update                 # Nao marcar tasks como concluidas automaticamente
--verbose                   # Tokens por fase + memory stats
--project <path>            # Projeto alvo (default: cwd)

Challenge Standalone

# Review adversarial do guardian output (requer OPENAI_API_KEY)
devflow challenge

# Passando arquivo especifico
devflow challenge docs/security/guardian-review.md

# Com spec para contexto
devflow challenge docs/security/guardian-review.md --spec docs/specs/feature.md

# Modelo (gpt-5.4 padrao, o3 para casos criticos)
devflow challenge --model o3

| Modelo | Custo/review | Quando usar | | --------- | ------------ | ----------------------------------------- | | gpt-5.4 | ~$0.01 | Padrao — rapido, code-focused | | o3-mini | ~$0.03 | Reasoning alternativo | | o3 | ~$0.28 | Seguranca critica, compliance, pagamentos |


Estrutura gerada no projeto

seu-projeto/
├── .claude/
│   ├── settings.json   # Agent Teams habilitado (automatico)
│   └── commands/
│       ├── agents/     # Definicoes dos 7 agentes + team
│       └── quick/      # Quick start commands
│
├── .devflow/
│   ├── agents/         # Metadata dos agentes
│   └── project.yaml    # Estado + versao do projeto
│
└── docs/
    ├── decisions/      # ADRs
    ├── planning/       # Stories e specs
    ├── snapshots/      # Historico do projeto
    └── system-design/  # SDDs, RFCs, capacity plans
        ├── sdd/
        ├── rfc/
        ├── capacity/
        └── trade-offs/

Versoes

| Versao | Features | | ------------- | ---------------------------------------------------------------------------------------- | | v0.1.0 | Multi-agent system, Documentation automation | | v0.2.0 | Structured metadata, Knowledge graph | | v0.3.0 | Hard stops, Mandatory delegation | | v0.4.0–v0.7.0 | Web IDE, System Designer agent, npm package | | v0.8.0 | Autopilot terminal-based, CLI commands | | v1.0.0 | Security hardening, npm global install fix | | v1.1.1 | Dual Scaling Modes: Parallel Subagents + Claude Agent Teams | | v1.2.0 | Remocao da Web IDE, Persistent Memory, Adaptive Autopilot, Context Isolation, Challenger | | v1.2.5 | Agent Teams habilitado por padrao no init/update | | v1.2.8 | Spawn templates com contexto obrigatorio para teammates | | v1.2.9 | Cores por agente, scores do challenger, artefatos persistidos, complexidade adaptativa | | v1.3.1 | Security fix SEG-01, /quick:refactor + /quick:debug, challenger.md reescrito, bug fixes |


Documentacao


Licenca

MIT License - veja LICENSE para detalhes.


DevFlow v1.3.1 - Desenvolvido por Evolve Labs