@aksp/opencrew 1.0.0 → 1.0.1

package/CHANGELOG.md

@@ -3,6 +3,15 @@
 All notable changes to opencrew are documented here.
 The format is based on [Keep a Changelog](https://keepachangelog.com/).
 
+## [1.0.1] — 2026-08-01
+
+### Changed
+- API keys for optional skills are now requested conversationally in chat (during crew
+  creation or skill install) instead of requiring the user to manually copy/edit `.env`
+  beforehand. Values are collected and written to `.env` automatically.
+- `init` no longer tells users to configure `.env` as a next step — no setup is required
+  to start using opencrew.
+
 ## [1.0.0] — 2026-08-01
 
 ### Added

package/README.md

@@ -1,61 +1,75 @@
 # opencrew
 
-**Create AI agent crews that work together — right from your IDE.**
+**Crie equipes de agentes de IA que trabalham juntos — direto na sua IDE.**
 
-opencrew is a multi-agent orchestration framework. Describe what you need in plain
-language and it assembles a team of specialized agents that run as an automated pipeline
-with human approval checkpoints. It works across Claude Code, Cursor, Codex, Gemini CLI,
-OpenCode, Antigravity and more.
+> 🇧🇷 This README is in Brazilian Portuguese (PT-BR), the project's primary audience. If
+> you don't read Portuguese, use your browser's translator (e.g. Google Translate) or ask
+> an AI assistant to translate this page.
 
-> **This is not an original framework — it's my personal build of OpenSquad.**
-> [OpenSquad](https://github.com/renatoasse/opensquad) was created by
+opencrew é um framework de orquestração multi-agente. Descreva o que você precisa em
+linguagem natural e ele monta um time de agentes especializados que rodam como um pipeline
+automatizado, com pontos de aprovação humana. Funciona no Claude Code, Cursor, Codex,
+Gemini CLI, OpenCode, Antigravity e mais.
+
+> **Este não é um framework original — é a minha versão pessoal do OpenSquad.**
+> O [OpenSquad](https://github.com/renatoasse/opensquad) foi criado por
 > [Renato Asse](https://github.com/renatoasse) ([Comunidade Sem Codar](https://semcodar.com.br)).
-> I ([aksp](https://www.npmjs.com/~aksp)) use OpenSquad daily and wanted to improve a few
-> things for my own workflow — so `opencrew` is that reworked version, shared in case it
-> helps others too. All credit for the original idea and framework goes to Renato Asse.
-> See [Origin & credits](#origin--credits) below. MIT licensed, same as the original.
+> Eu ([aksp](https://www.npmjs.com/~aksp)) uso o OpenSquad no dia a dia e quis melhorar
+> algumas coisas para o meu próprio fluxo de trabalho — o `opencrew` é essa versão
+> reformulada, compartilhada caso ajude outras pessoas também. Todo o crédito pela ideia e
+> pelo framework original é do Renato Asse. Veja [Origem e créditos](#origem-e-créditos)
+> abaixo. Licenciado sob MIT, assim como o original.
 
 ---
 
-## Installation
+## Instalação
 
-**Prerequisite:** Node.js 20+
+**Pré-requisito:** Node.js 20+
 
 ```bash
 npx @aksp/opencrew init
 ```
 
-`init` scaffolds the workspace in the current folder and asks which AI IDEs you use, then
-generates the right integration files for each. Then:
+O `init` monta o workspace na pasta atual e pergunta quais IDEs de IA você usa, gerando os
+arquivos de integração certos para cada uma. Depois:
+
+1. Abra a pasta na sua IDE de IA.
+2. Digite `/opencrew` para começar — a primeira execução configura o perfil da sua empresa.
+
+Isso já é suficiente para criar e rodar crews: nenhuma configuração prévia, instalação
+extra ou chave de API é necessária para começar.
 
-1. Open the folder in your AI IDE.
-2. Copy `.env.example` to `.env` and fill in keys for the skills you plan to use.
-3. Type `/opencrew` to start — the first run sets up your company profile.
+Algumas **skills opcionais** dependem de serviços externos — por exemplo, publicar no
+Instagram, gerar imagens com IA, fazer web scraping (Apify) ou enviar e-mails (Resend). Você
+não precisa se preocupar com isso antes de começar: se, ao montar uma crew, o opencrew
+identificar que ela precisa de uma dessas skills, ele pede a chave direto na conversa
+(explicando o que é e onde consegui-la) e salva tudo por conta própria. Não é preciso abrir
+nem editar nenhum arquivo manualmente.
 
-Preselect IDEs (skip the prompt) or configure all of them:
+Você pode pré-selecionar as IDEs (pula a pergunta) ou configurar todas de uma vez:
 
 ```bash
 npx @aksp/opencrew init --ide=claude-code,codex
 npx @aksp/opencrew init --all
 ```
 
-## Updating
+## Atualizando
 
-Refresh the framework without losing your work:
+Atualize o framework sem perder o seu trabalho:
 
 ```bash
 npx @aksp/opencrew update
 ```
 
-`update` refreshes only `_opencrew/core`, the catalog skills and `AGENTS.md`.
-Your `crews/`, memory, IDE bridges and `.env` are left untouched.
+O `update` atualiza apenas `_opencrew/core`, as skills do catálogo e o `AGENTS.md`. Suas
+`crews/`, memória, integrações de IDE e `.env` continuam intactos.
 
-## Supported IDEs
+## IDEs suportadas
 
-| IDE | Bridge file(s) generated |
+| IDE | Arquivo(s) de integração gerado(s) |
 |-----|--------------------------|
 | Claude Code | `.claude/skills/opencrew/SKILL.md`, `CLAUDE.md` |
-| Codex (OpenAI) | `AGENTS.md` (native) + `.agents/skills/opencrew/SKILL.md` |
+| Codex (OpenAI) | `AGENTS.md` (nativo) + `.agents/skills/opencrew/SKILL.md` |
 | Cursor | `.cursor/rules/opencrew.mdc` |
 | VS Code + Copilot | `.github/copilot-instructions.md` |
 | OpenCode | `.opencode/commands/opencrew.md` |
@@ -64,55 +78,57 @@ Your `crews/`, memory, IDE bridges and `.env` are left untouched.
 | Qwen Code | `QWEN.md` |
 | Trae | `.trae/rules/opencrew.md` |
 
-Every bridge is a thin pointer to the single source of truth, **`AGENTS.md`**.
+Cada integração é só um ponteiro enxuto para a fonte única de verdade, o **`AGENTS.md`**.
 
-## How it works
+## Como funciona
 
-- **Architect** designs a crew from your description (agents, pipeline, skills).
-- **Sherlock** (optional) analyzes reference profiles to extract real content patterns.
-- **Pipeline Runner** executes the crew, pausing at checkpoints for your approval.
-- **Skills Engine** loads integrations (scraping, design, publishing, email…) on demand,
-  using a two-tier scheme to keep token usage low.
+- **Architect** projeta uma crew a partir da sua descrição (agentes, pipeline, skills).
+- **Sherlock** (opcional) analisa perfis de referência para extrair padrões reais de conteúdo.
+- **Pipeline Runner** executa a crew, pausando nos checkpoints para sua aprovação.
+- **Skills Engine** carrega integrações (scraping, design, publicação, e-mail…) sob demanda,
+  usando um esquema em duas camadas para manter o consumo de tokens baixo.
 
-## Commands (inside your IDE)
+## Comandos (dentro da sua IDE)
 
-| Command | What it does |
+| Comando | O que faz |
 |---------|--------------|
-| `/opencrew` | Open the main menu |
-| `/opencrew create <description>` | Create a new crew |
-| `/opencrew run <name>` | Run a crew |
-| `/opencrew list` | List your crews |
-| `/opencrew edit <name>` | Modify a crew |
-| `/opencrew skills` | Browse / install / remove skills |
-
-## For maintainers
-
-See [CONTRIBUTING.md](CONTRIBUTING.md). The golden rule: `AGENTS.md` is the only place
-system instructions live; per-IDE files are generated from `src/lib/ides.js`.
-
-## Origin & credits
-
-`opencrew` is a **reworked distribution of [OpenSquad](https://github.com/renatoasse/opensquad)**,
-the multi-agent orchestration framework created and maintained by
-**[Renato Asse](https://github.com/renatoasse)**, founder of
-[Comunidade Sem Codar](https://semcodar.com.br). The original project, its concept, agent
-model, pipeline design and skill system are his work — please star and follow the
-[upstream repository](https://github.com/renatoasse/opensquad) and watch the
-[launch video](https://www.youtube.com/watch?v=CL1ppI4qHeU).
-
-I use OpenSquad in my day-to-day and made a few changes that fit how I work, which I believe
-can help other people as well. What this build changes relative to upstream:
-
-- **npm installer under my scope** — `npx @aksp/opencrew init` / `update`, with a
-  non-destructive update path that preserves your crews, memory and `.env`.
-- **Single source of truth for multi-IDE** — one canonical `AGENTS.md`; every IDE gets a
-  thin generated bridge file instead of a hand-maintained document per tool.
-- **Token-economy and skill-invocation improvements** to the framework core.
-
-This is an independent, community fork — it is **not** affiliated with or endorsed by
-Renato Asse or Comunidade Sem Codar. If you want the official project, use
+| `/opencrew` | Abre o menu principal |
+| `/opencrew create <descrição>` | Cria uma nova crew |
+| `/opencrew run <nome>` | Executa uma crew |
+| `/opencrew list` | Lista suas crews |
+| `/opencrew edit <nome>` | Modifica uma crew |
+| `/opencrew skills` | Navega / instala / remove skills |
+
+## Para mantenedores
+
+Veja [CONTRIBUTING.md](CONTRIBUTING.md). Regra de ouro: `AGENTS.md` é o único lugar onde
+vivem as instruções do sistema; os arquivos por IDE são gerados a partir de `src/lib/ides.js`.
+
+## Origem e créditos
+
+`opencrew` é uma **distribuição reformulada do [OpenSquad](https://github.com/renatoasse/opensquad)**,
+o framework de orquestração multi-agente criado e mantido por
+**[Renato Asse](https://github.com/renatoasse)**, fundador da
+[Comunidade Sem Codar](https://semcodar.com.br). O projeto original, seu conceito, o modelo
+de agentes, o design do pipeline e o sistema de skills são trabalho dele — dê uma estrela e
+siga o [repositório original](https://github.com/renatoasse/opensquad) e assista ao
+[vídeo de lançamento](https://www.youtube.com/watch?v=CL1ppI4qHeU).
+
+Eu uso o OpenSquad no meu dia a dia e fiz algumas mudanças que se encaixam no meu jeito de
+trabalhar, que acredito que também podem ajudar outras pessoas. O que esta versão muda em
+relação ao original:
+
+- **Instalador npm no meu escopo** — `npx @aksp/opencrew init` / `update`, com um caminho
+  de atualização não destrutivo que preserva suas crews, memória e `.env`.
+- **Fonte única de verdade para multi-IDE** — um `AGENTS.md` canônico; cada IDE recebe um
+  arquivo de integração gerado e enxuto, em vez de um documento mantido manualmente para
+  cada ferramenta.
+- **Melhorias de economia de tokens e de invocação de skills** no núcleo do framework.
+
+Este é um fork independente, feito pela comunidade — **não** é afiliado nem endossado pelo
+Renato Asse ou pela Comunidade Sem Codar. Se você quiser o projeto oficial, use
 [`npx opensquad init`](https://github.com/renatoasse/opensquad).
 
-## License
+## Licença
 
-MIT — see [LICENSE](LICENSE). Original OpenSquad framework © Renato Asse, also MIT.
+MIT — veja [LICENSE](LICENSE). Framework OpenSquad original © Renato Asse, também MIT.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@aksp/opencrew",
-  "version": "1.0.0",
+  "version": "1.0.1",
   "publishConfig": {
     "access": "public"
   },

package/src/commands/init.js

@@ -82,8 +82,8 @@ export async function init(opts = {}) {
   log(`\n${c.green(c.bold('Done!'))} opencrew is installed.\n`);
   log(`${c.bold('Next steps:')}`);
   log(`  1. Open this folder in your AI IDE.`);
-  log(`  2. Copy ${c.cyan('.env.example')} to ${c.cyan('.env')} and fill in keys for the skills you use.`);
-  log(`  3. Type ${c.cyan('/opencrew')} to start (first run sets up your company profile).\n`);
+  log(`  2. Type ${c.cyan('/opencrew')} to start (first run sets up your company profile).`);
+  log(`     No API keys needed up front — opencrew asks for them in chat only if a skill you use requires one.\n`);
 }
 
 async function tpl(name) {

package/templates/_opencrew/.opencrew-version

@@ -1 +1 @@
-0.1.15
+1.0.1

package/templates/_opencrew/core/skills.engine.md

@@ -121,10 +121,25 @@ For the full SKILL.md specification, see `skills/opencrew-skill-creator/referenc
 
    #### a. Environment Variables (if `env:` is present)
 
-   For each variable listed in the `env` array, check if it exists in the project `.env` file.
-   - For each missing variable, inform the user:
-     "⚠️ Environment variable `{VAR_NAME}` is required by {skill name}. Add it to your `.env` file."
-   - Do NOT block installation — warn only. The skill is installed even without env vars configured.
+   This is a conversational step — never ask the user to open or edit `.env` manually.
+   The audience is non-technical (managers, analysts), so all secrets are collected in
+   chat and written to disk automatically.
+
+   For each variable listed in the `env` array:
+   1. Check if it already exists (non-empty) in the project `.env` file → if so, reuse it
+      silently, skip to the next variable.
+   2. If missing, ask the user directly in chat:
+      "This skill needs an API key to work: **{VAR_NAME}** ({one-line purpose, e.g. 'your
+      Instagram access token'}). Paste it here, or press Enter to skip and set it up later."
+   3. If the user provides a value → write it to the project `.env` file (create the file
+      from `.env.example` if it doesn't exist yet). Confirm briefly: "✅ Saved."
+   4. If the user skips → inform them:
+      "⚠️ {skill name} won't fully work until `{VAR_NAME}` is set. You can add it anytime —
+      just ask me to configure {skill name} again."
+   - Do NOT block installation either way — the skill is installed regardless of whether
+     env vars were provided.
+   - Values collected here are reused in steps b/c below — do not ask for the same
+     variable twice.
 
    #### b. MCP Setup — stdio transport (if `type: mcp` or `type: hybrid` with `mcp.transport: stdio` or no transport specified)
 
@@ -135,9 +150,8 @@ For the full SKILL.md specification, see `skills/opencrew-skill-creator/referenc
         1. Yes, overwrite
         2. No, keep existing
         If "No" → skip MCP configuration but still complete installation
-   3. For each env var listed in the skill's `env` array:
-      - Check `.env` for an existing value
-      - If missing, ask the user to type the value directly
+   3. Use the values already collected in step 6.a for each env var in the skill's `env`
+      array (do not ask again).
    4. Add to `mcpServers`:
       ```json
       "{server_name}": {
@@ -154,9 +168,8 @@ For the full SKILL.md specification, see `skills/opencrew-skill-creator/referenc
 
    1. Read `.claude/settings.local.json` (create with `{"mcpServers": {}}` if it doesn't exist)
    2. Check for `server_name` conflict (same as stdio above)
-   3. For each env var listed in the skill's `env` array:
-      - Check `.env` for an existing value
-      - If missing, ask the user to type the value directly
+   3. Use the values already collected in step 6.a for each env var in the skill's `env`
+      array (do not ask again).
    4. Build the mcpServers entry:
       - Start with `{ "type": "http", "url": "{url}" }`
       - If the skill has a `headers` field in `mcp`, add a `"headers"` object by resolving
@@ -267,8 +280,11 @@ before the pipeline begins (fail fast).
 
    d. **Verify env vars** (if `env:` is present):
       - Check each variable in `.env`
-      - If any are missing → warn the user but do NOT block pipeline execution.
-        "⚠️ Skill '{skill}' is missing environment variable(s): {list}. It may not work correctly."
+      - If any are missing → ask the user conversationally, right here in chat, the same
+        way as Operation 2, step 6.a (never point them to the `.env` file to edit it
+        themselves). If they provide a value, write it to `.env` and continue. If they
+        skip, warn: "⚠️ Skill '{skill}' is missing environment variable(s): {list}. It may
+        not work correctly." — but do NOT block pipeline execution either way.
 
 4. **Return resolved skill list**: Return all resolved skills with their parsed frontmatter
    and SKILL.md body content. This list is used by Operation 6 to inject instructions.