wize-dev-kit 0.1.0 → 0.1.2

package/CHANGELOG.md

@@ -5,6 +5,43 @@ Format inspired by [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 
 ## [Unreleased]
 
+## [0.1.2] — 2026-05-31
+
+### Added
+
+- **Claude Code adapter is now functional.** `npx wize-dev-kit install` (when `claude-code` is selected as IDE target) generates `.claude/skills/wize-*/SKILL.md` for every agent, workflow and skill in the kit. Each SKILL.md uses the Claude Code skill frontmatter so the slash menu picks up `/wize-orchestrator`, `/wize-product-brief`, `/wize-create-prd`, `/wize-tea-gate`, etc.
+- **`wize-help` reworked as an orchestrator-aware skill.** Moved from `core-skills/` to `orchestrator-skills/`. The skill instructs Wizer to read `.wize/config/project.toml` plus the state of `.wize/planning/`, `.wize/solutioning/`, `.wize/implementation/` and reply with a phase-aware "what to do next" answer. Four modes: default, `next`, `status`, `personas`.
+- **Interactive prompts in the installer.** Profile and IDE-target selection now use arrow-keys + space (checkbox) via the `prompts` dependency. Non-TTY environments fall back to the previous number-based prompt automatically (preserves CI scriptability).
+- Adapter test suite (`test/adapter-claude-code.test.js`) covering: per-agent SKILL.md emission, overlay-gate behavior (core-only vs full), `wize-help` content checks, dry-run safety.
+
+### Changed
+
+- Profile gating now lives inside the Claude Code adapter: overlay workflows (`wize-web-*`, `wize-app-*`) are only emitted when the corresponding overlay is in the selected profiles.
+
+### Dependencies
+
+- Added `prompts@^2.4.2` (1 sub-dependency, ~5 KB). Only used by the interactive installer; not bundled into the kit's runtime artifacts.
+
+## [0.1.1] — 2026-05-31
+
+### Changed
+
+- `README.md`: `Install` section now appears right after the badges (was buried in the middle).
+- `ROSTER.md`: rewritten entirely in English; Wizer's motto no longer references a personal name — now reads "I know the qwize methodology, I know the project — I activate the right agent."
+- Installer: now asks **two** language questions separately (BMAD parity):
+  - **Communication language** — how agents talk to you in chat.
+  - **Document output language** — language used in generated files (`brief.md`, `prd.md`, `architecture.md`, gates, etc.).
+  - Both stored under `[language]` in `.wize/config/project.toml`.
+  - Includes a curated catalog of common BCP-47 codes (en, pt-BR, pt-PT, es, fr, de, it, zh-CN, ja, vi) with free-text fallback for any other locale.
+- `package.json`: corrected `repository`, `homepage`, `bugs` to point at `qwize-br/wize-development-kit` (was `qwize/wize-dev-kit`).
+- `package.json`: added `prepublishOnly` running tests + structural validators.
+
+### Added
+
+- `.github/workflows/publish.yml`: GitHub Actions workflow that publishes to npm on tag push (`v*`) using **Trusted Publishing** (OIDC). No long-lived `NPM_TOKEN` secret. Emits `--provenance` attestation.
+
+## [0.1.0] — 2026-05-31
+
 ### Added
 
 - Initial skeleton (v0.1.0 scaffold).
@@ -21,4 +58,7 @@ Format inspired by [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 - Inspired by [BMAD Method v6.8.0](https://github.com/bmad-code-org/BMAD-METHOD).
 - WDS module inspired by [bmad-method-wds-expansion](https://github.com/bmad-code-org/bmad-method-wds-expansion).
 
-[Unreleased]: https://github.com/qwize/wize-dev-kit/compare/v0.1.0...HEAD
+[Unreleased]: https://github.com/qwize-br/wize-development-kit/compare/v0.1.2...HEAD
+[0.1.2]: https://github.com/qwize-br/wize-development-kit/compare/v0.1.1...v0.1.2
+[0.1.1]: https://github.com/qwize-br/wize-development-kit/compare/v0.1.0...v0.1.1
+[0.1.0]: https://github.com/qwize-br/wize-development-kit/releases/tag/v0.1.0

package/README.md

@@ -10,6 +10,36 @@
 
 ---
 
+## Install
+
+In any greenfield or brownfield repo:
+
+```bash
+npx wize-dev-kit install
+```
+
+Or straight from GitHub (no npm required):
+
+```bash
+npx github:qwize-br/wize-development-kit install
+```
+
+The installer asks:
+
+1. Profile(s) to enable (Core / +Web / +App — multi-select).
+2. IDE target(s) (Claude Code, Cursor, Windsurf, Codex, Continue, Kimi Code, OpenCode, Antigravity, or generic fallback — multi-select).
+3. Language preferences (communication + document output).
+4. Output folder (default `.wize/`).
+5. For brownfield repos: offers to run `wize-document-project` to baseline the existing codebase.
+
+After install, open your IDE and start with:
+
+> "Activate Wizer and brief him on the project."
+
+Wizer will route you to the right persona (Pepper for brief, Mantis for UX baseline, Tony for architecture preferences, etc.).
+
+---
+
 ## What it is
 
 Wize Development Kit (WDK) is an installable AI agent stack that takes a project from **brief → PRD → UX strategy → architecture → tested implementation** through 9 specialized agents (Marvel-themed) and a structured workflow library. It runs inside your AI IDE (Claude Code, Cursor, Windsurf, and others) and writes structured artifacts to a hidden `.wize/` folder in the target repo.
@@ -42,35 +72,6 @@ See [`ROSTER.md`](ROSTER.md) for personas, styles and BMAD equivalences.
 
 ---
 
-## Quick start
-
-```bash
-# In any greenfield or brownfield repo
-npx wize-dev-kit install
-```
-
-Or, before this is published in your environment, install straight from GitHub:
-
-```bash
-npx github:qwize-br/wize-development-kit install
-```
-
-The installer asks:
-
-1. Profile(s) to enable (Core / +Web / +App — multi-select).
-2. IDE target(s) (Claude Code, Cursor, Windsurf, Codex, Continue, Kimi Code, OpenCode, Antigravity, or generic fallback — multi-select).
-3. Language preferences (communication + document output).
-4. Output folder (default `.wize/`).
-5. For brownfield repos: offers to run `wize-document-project` to baseline the existing codebase.
-
-After install, open your IDE and start with:
-
-> "Activate Wizer and brief him on the project."
-
-Wizer will route you to the right persona (Pepper for brief, Mantis for UX baseline, Tony for architecture preferences, etc.).
-
----
-
 ## Output layout (in the target repo)
 
 ```

package/ROSTER.md

@@ -1,25 +1,25 @@
-# Wize Dev Kit — Roster de Agentes (proposta v1)
+# Wize Dev Kit — Agent Roster (v1)
 
-> Tema: **Marvel**. Cada persona representa um papel específico no método.
-> Equivalências BMAD listadas para referência cruzada com o repositório-pai.
+> Theme: **Marvel**. Each persona represents a specific role in the qwize methodology.
+> BMAD equivalences listed for cross-reference with the inspiration repo.
 
-## Quadro
+## Roster table
 
-| # | Persona | Code | Papel | Fase | Lema / Estilo | Equiv. BMAD |
+| # | Persona | Code | Role | Phase | Motto / Style | BMAD equiv. |
 |---|---|---|---|---|---|---|
-| 1 | **Wizer** | `wize-orchestrator` | Orchestrator / Knowledge Base / Briefing | Todas | "Conheço o time, conheço o André, ative o agente certo." Estilo: anfitrião, escolhe, distribui. Tools: globais. | (não existe — papel de orchestrator surge no `party-mode` do BMAD) |
-| 2 | **Pepper Potts** | `wize-agent-analyst` | Business Analyst | 1 — Analysis | "Eficiência implacável." Brainstorming, market research, product-brief, PRFAQ, ROI, stakeholder map. Estilo: pragmática, antecipa, conecta business → tech. | Mary (`bmad-agent-analyst`) |
-| 3 | **Peggy Carter** | `wize-agent-tech-writer` | Technical Writer | Transversal (1–4) | "Estrutura, audiência, clareza." DITA, CommonMark, OpenAPI, READMEs, runbooks. Estilo: organizada, didática, técnica-mas-acessível. | Paige (`bmad-agent-tech-writer`) |
-| 4 | **Maria Hill** | `wize-agent-pm` | Product Manager | 2 — Planning | "Mission first." PRD, epics, sprint planning, cobrança de prazos. Estilo: militar, disciplinado, sem desculpas. | John (`bmad-agent-pm`) |
-| 5 | **Mantis** | `wize-agent-ux-designer` | UX Designer (Whiteport Strategic UX v0.4.3) | 2–3 | "Sinto o usuário antes de desenhar." Jobs-to-be-done, jornadas, empathy mapping, design tokens, IA, design system. Estilo: research-heavy, qualitative, narrativa empática. | Sally (`bmad-agent-ux-designer`) — substituída + metodologia Whiteport |
-| 6 | **Nick Fury** | `wize-agent-solution-strategist` | Solution Strategy / Tech Vision | 2 → 3 (boundary) | "Pessoas > Objetivo." Big-picture, NFRs, escolha de stack, princípios, trade-offs estratégicos. Estilo: autoritário, direto, poucas palavras. | (parcial de) Winston — vertente estratégica |
-| 7 | **Tony Stark** | `wize-agent-architect` | System Architect | 3 — Solutioning | "Eu construo as coisas." System design, componentes, ADRs, prototipagem, escolha de patterns. Estilo: confiante, irreverente, mostra com código. | Winston (`bmad-agent-architect`) — vertente sistêmica |
-| 8 | **Hawkeye** | `wize-agent-test-architect` | Test Architect (TEA) | Transversal (gates em 2, 3, 4) | "Sempre acerto onde dói." Risk profile, test design, traceability, NFR assessment, review, gate decision. Estilo: pragmático, edge-case hunter, foca no que importa. | (novo — não existe no BMAD, inspirado no TEA do BMAD-Method v5) |
-| 9 | **Shuri** | `wize-agent-dev` | Senior Developer | 4 — Implementation | "Wakanda forever — agora compila." TDD red-green-refactor, segurança, performance. Estilo: gênio inovador, rápida, protetora do ecossistema, código limpo. | Amelia (`bmad-agent-dev`) |
+| 1 | **Wizer** | `wize-orchestrator` | Orchestrator / Knowledge Base / Briefing | All | "I know the qwize methodology, I know the project — I activate the right agent." Style: host, picks, routes. Tools: global. | (none — orchestrator role surfaces only in BMAD's `party-mode`) |
+| 2 | **Pepper Potts** | `wize-agent-analyst` | Business Analyst | 1 — Analysis | "Relentless efficiency." Brainstorming, market research, product brief, PR/FAQ, ROI, stakeholder map. Style: pragmatic, anticipates, connects business → tech. | Mary (`bmad-agent-analyst`) |
+| 3 | **Peggy Carter** | `wize-agent-tech-writer` | Technical Writer | Transversal (1–4) | "Structure, audience, clarity." DITA, CommonMark, OpenAPI, READMEs, runbooks. Style: organized, didactic, technical but accessible. | Paige (`bmad-agent-tech-writer`) |
+| 4 | **Maria Hill** | `wize-agent-pm` | Product Manager | 2 — Planning | "Mission first." PRD, epics, sprint planning, deadline enforcement. Style: military discipline, no excuses, outcome-focused. | John (`bmad-agent-pm`) |
+| 5 | **Mantis** | `wize-agent-ux-designer` | UX Designer (Whiteport Strategic UX v0.4.3) | 2–3 | "I feel the user before I sketch." Jobs-to-be-done, journeys, empathy mapping, design tokens, IA, design system. Style: research-heavy, qualitative, empathic narrative. | Sally (`bmad-agent-ux-designer`) — replaced with Whiteport methodology |
+| 6 | **Nick Fury** | `wize-agent-solution-strategist` | Solution Strategy / Tech Vision | 2 → 3 (boundary) | "People > Objective." Big-picture, NFRs, stack family, principles, strategic trade-offs. Style: authoritative, direct, few words. | (partial) Winston — strategic side |
+| 7 | **Tony Stark** | `wize-agent-architect` | System Architect | 3 — Solutioning | "I build the thing." System design, components, ADRs, prototyping, pattern selection. Style: confident, irreverent, shows with code. | Winston (`bmad-agent-architect`) — systemic side |
+| 8 | **Hawkeye** | `wize-agent-test-architect` | Test Architect (TEA) | Transversal (gates in 2, 3, 4) | "I always hit where it hurts." Risk profile, test design, traceability, NFR assessment, review, gate decision. Style: pragmatic, edge-case hunter, focused on what matters. | (new — does not exist in BMAD core; inspired by BMAD-Method v5 TEA) |
+| 9 | **Shuri** | `wize-agent-dev` | Senior Developer | 4 — Implementation | "Wakanda forever — now it compiles." TDD red-green-refactor, security, performance. Style: genius innovator, fast, clean code, protective of the ecosystem. | Amelia (`bmad-agent-dev`) |
 
-## Notas
+## Notes
 
-- **Tema visual:** ícones podem usar emojis (📊 🛡️ 🦾 etc.) mas a identidade visual fica para a Fase 8 de consolidação.
-- **Agent Builder:** decidido na Fase 3 como **skill**, não agente (`wize-create-agent`). Wizer chama essa skill quando precisa registrar um novo persona/módulo custom.
-- **Times:** todos no time default `software-development`. Times opcionais (`web-dev`, `app-dev`) ativam variantes de Tony e Shuri por perfil (decisão pendente para Fase 4).
-- **Adições futuras (fora do kit dev):** Pepper já está usada como Analyst; Fury já como Strategist. Outras personas do ecossistema Marvel (Black Panther, Wanda, Falcon, Vision, Riri Williams) ficam livres para kits futuros (Wize Ops Kit, Wize Data Kit, Wize Security Kit, etc.).
+- **Visual theme:** emoji icons (📊 🛡️ 🦾 etc.) are placeholders; a full visual identity will be consolidated in a later phase.
+- **Agent Builder:** decided as a **skill** (`wize-create-agent`), not an agent. Wizer invokes the skill when a new persona or custom module needs to be registered.
+- **Teams:** default team is `software-development`. Optional teams (`web-dev`, `app-dev`) activate variants of Tony and Shuri per profile.
+- **Future Marvel personas (out of dev-kit scope):** Pepper is already Analyst here, Fury is already Strategist. Other Marvel personas (Black Panther, Wanda, Falcon, Vision, Riri Williams, Kamala Khan) remain reserved for future kits (Wize Ops Kit, Wize Data Kit, Wize Security Kit, etc.).

package/adapters/claude-code/render.js

@@ -1,15 +1,186 @@
-/* Stub renderer for this adapter — emits a console preview of what would be written. */
+//
+// Claude Code adapter — renders kit assets into .claude/skills/wize-{code}/SKILL.md.
+//
+// Strategy:
+//   - For each agent (agent.yaml + persona.md), emit a SKILL.md that pairs the
+//     agent identity with its responsibilities, so Claude Code can activate
+//     the persona via slash command.
+//   - For each workflow/skill file, emit a SKILL.md that wraps the workflow
+//     content with an activation frontmatter Claude Code understands.
+//
+// Each SKILL.md uses the Claude Code skill format:
+//     ---
+//     name: <code>
+//     description: <one-liner shown in the slash menu>
+//     ---
+//     <body>
+//
 'use strict';
 
 const fs = require('node:fs');
 const path = require('node:path');
+const { walkAgents, walkWorkflows, walkSkills } = require('../../tools/installer/validators/walk.js');
 
+function readYamlField(content, field) {
+  // Minimal scalar/quoted-string extractor for top-level YAML scalars.
+  const re = new RegExp(`^${field}:\\s*(?:"([^"]*)"|'([^']*)'|(.*?))\\s*$`, 'm');
+  const m = content.match(re);
+  if (!m) return null;
+  return (m[1] || m[2] || m[3] || '').trim();
+}
+
+function readFrontmatter(content) {
+  // Returns an object of top-level key:value pairs from a leading
+  // ---\n...\n--- block. Simple scalars only.
+  if (!content.startsWith('---')) return {};
+  const end = content.indexOf('\n---', 3);
+  if (end === -1) return {};
+  const fm = content.slice(3, end).split('\n');
+  const out = {};
+  for (const line of fm) {
+    const m = line.match(/^([a-zA-Z_][a-zA-Z0-9_-]*):\s*(.*?)\s*$/);
+    if (m) out[m[1]] = m[2].replace(/^"|"$/g, '').replace(/^'|'$/g, '');
+  }
+  return out;
+}
+
+function bodyAfterFrontmatter(content) {
+  if (!content.startsWith('---')) return content;
+  const end = content.indexOf('\n---', 3);
+  if (end === -1) return content;
+  return content.slice(end + 4).replace(/^\s+/, '');
+}
+
+function escapeForYamlDouble(str) {
+  return String(str).replace(/\\/g, '\\\\').replace(/"/g, '\\"');
+}
+
+function clipOneLine(str, max = 240) {
+  const flat = String(str).replace(/\s+/g, ' ').trim();
+  if (flat.length <= max) return flat;
+  return flat.slice(0, max - 1).trimEnd() + '…';
+}
+
+function renderAgentSkill({ code, name, title, description, body }) {
+  const desc = clipOneLine(`${name} (${title}) — ${description}`);
+  return `---
+name: ${code}
+description: "${escapeForYamlDouble(desc)}"
+---
+
+# ${name} — ${title}
+
+${body.trim()}
+`;
+}
+
+function renderWorkflowSkill({ code, name, description, body }) {
+  const desc = clipOneLine(description || `Workflow ${code}`);
+  return `---
+name: ${code}
+description: "${escapeForYamlDouble(desc)}"
+---
+
+# ${name}
+
+${body.trim()}
+`;
+}
+
+function ensureDir(dir) {
+  fs.mkdirSync(dir, { recursive: true });
+}
+
+function writeSkill(skillsDir, code, content) {
+  const dir = path.join(skillsDir, code);
+  ensureDir(dir);
+  fs.writeFileSync(path.join(dir, 'SKILL.md'), content, 'utf-8');
+}
+
+/**
+ * Render the full kit into `.claude/skills/` at the project root.
+ *
+ * @param {string} kitRoot - absolute path to the installed wize-dev-kit package
+ * @param {string} projectRoot - absolute path to the target repo (cwd of install)
+ * @param {object} opts
+ * @param {string[]} [opts.profiles] - selected profile codes (e.g., ['core', 'web-overlay'])
+ * @param {boolean} [opts.dryRun] - when true, return the list without writing
+ * @returns {{ written: string[], skipped: string[] }}
+ */
 function render(kitRoot, projectRoot, opts = {}) {
-  const adapterDir = __dirname;
-  const cfg = require('node:fs').readFileSync(path.join(adapterDir, 'adapter.yaml'), 'utf-8');
-  const targetPath = (cfg.match(/^target_path:\s*"(.+)"$/m) || [])[1] || '';
-  const filePattern = (cfg.match(/^file_pattern:\s*"(.+)"$/m) || [])[1] || '';
-  console.log(`[adapter:${path.basename(adapterDir)}] would emit ${filePattern} under ${path.join(projectRoot, targetPath)}`);
+  const skillsDir = path.join(projectRoot, '.claude', 'skills');
+  const written = [];
+  const skipped = [];
+  const profiles = new Set(opts.profiles || ['core']);
+  const dryRun = !!opts.dryRun;
+
+  if (!dryRun) ensureDir(skillsDir);
+
+  // 1. Agents
+  for (const yamlPath of walkAgents(kitRoot)) {
+    const dir = path.dirname(yamlPath);
+    const yaml = fs.readFileSync(yamlPath, 'utf-8');
+    const code = readYamlField(yaml, 'code');
+    const name = readYamlField(yaml, 'name');
+    const title = readYamlField(yaml, 'title');
+    const description = readYamlField(yaml, 'description') || '';
+    if (!code || !name) { skipped.push(yamlPath); continue; }
+
+    const personaPath = path.join(dir, 'persona.md');
+    const persona = fs.existsSync(personaPath) ? fs.readFileSync(personaPath, 'utf-8') : '';
+    const body = persona || description;
+
+    const content = renderAgentSkill({ code, name, title, description, body });
+    if (dryRun) {
+      written.push(`[dry-run] ${code}/SKILL.md`);
+    } else {
+      writeSkill(skillsDir, code, content);
+      written.push(`${code}/SKILL.md`);
+    }
+  }
+
+  // 2. Workflows — gate by overlay selection
+  for (const wfPath of walkWorkflows(kitRoot)) {
+    const content = fs.readFileSync(wfPath, 'utf-8');
+    const fm = readFrontmatter(content);
+    const code = fm.code;
+    const name = fm.name || code;
+    if (!code) { skipped.push(wfPath); continue; }
+
+    if (fm.overlay === 'web' && !profiles.has('web-overlay')) { skipped.push(wfPath); continue; }
+    if (fm.overlay === 'app' && !profiles.has('app-overlay')) { skipped.push(wfPath); continue; }
+
+    const description = `${fm.phase || fm.gate || 'workflow'}: ${name}`;
+    const body = bodyAfterFrontmatter(content);
+    const skillContent = renderWorkflowSkill({ code, name, description, body });
+    if (dryRun) {
+      written.push(`[dry-run] ${code}/SKILL.md`);
+    } else {
+      writeSkill(skillsDir, code, skillContent);
+      written.push(`${code}/SKILL.md`);
+    }
+  }
+
+  // 3. Standalone skills
+  for (const skPath of walkSkills(kitRoot)) {
+    const content = fs.readFileSync(skPath, 'utf-8');
+    const fm = readFrontmatter(content);
+    const code = fm.code;
+    const name = fm.name || code;
+    if (!code) { skipped.push(skPath); continue; }
+
+    const description = fm.module ? `${fm.module} skill: ${name}` : name;
+    const body = bodyAfterFrontmatter(content);
+    const skillContent = renderWorkflowSkill({ code, name, description, body });
+    if (dryRun) {
+      written.push(`[dry-run] ${code}/SKILL.md`);
+    } else {
+      writeSkill(skillsDir, code, skillContent);
+      written.push(`${code}/SKILL.md`);
+    }
+  }
+
+  return { written, skipped };
 }
 
 module.exports = { render };

package/package.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://json.schemastore.org/package.json",
   "name": "wize-dev-kit",
-  "version": "0.1.0",
+  "version": "0.1.2",
   "description": "Full-lifecycle AI-assisted development kit with Test Architect and Whiteport Design Studio embedded. Inspired by BMAD Method and WDS.",
   "keywords": [
     "ai",
@@ -48,5 +48,8 @@
     "ARCH.md",
     "ROSTER.md",
     "DECISIONS.md"
-  ]
+  ],
+  "dependencies": {
+    "prompts": "^2.4.2"
+  }
 }

package/src/core-skills/module.yaml

@@ -16,9 +16,6 @@ skills:
   - code: wize-brainstorming
     name: "Brainstorming"
     description: "Divergent-then-convergent ideation; produces idea-pool.md."
-  - code: wize-help
-    name: "Help"
-    description: "Routes user to the right agent/workflow based on intent."
   - code: wize-shard-doc
     name: "Shard Doc"
     description: "Splits large markdown docs (PRD, architecture) into addressable shards."

package/src/orchestrator-skills/module.yaml

@@ -13,6 +13,9 @@ agents:
     description: "Knows the user deeply, parses raw demand, routes to the right specialist. Speaks like a thoughtful host: warm welcome, sharp question, clear handoff."
 
 skills:
+  - code: wize-help
+    name: "Help (Wizer)"
+    description: "Project-state-aware help. Reads .wize/ and tells the user what to do next. Modes: default, next, status, personas."
   - code: wize-onboarding
     name: "Onboarding"
     description: "Post-install triage: greenfield vs brownfield, profile, objective. Delegates to specialists."

package/src/orchestrator-skills/wize-help/skill.md

@@ -0,0 +1,102 @@
+---
+code: wize-help
+name: Help (Wizer)
+module: orchestrator
+owner: wize-orchestrator
+status: ready
+---
+
+# Wizer · `/wize-help`
+
+You are **Wizer**, the orchestrator. The user invoked `/wize-help`. Don't dump a menu — read the project's state and give a short, actionable answer in the user's voice.
+
+## Modes
+
+| Invocation | What to do |
+|---|---|
+| `/wize-help` or `/wize` (no argument) | Greeting + project snapshot + the **single best next step**. |
+| `/wize-help next` | Just the next step. Skip the snapshot. |
+| `/wize-help status` | Snapshot only — phase, last TEA gate, in-flight stories. |
+| `/wize-help personas` | List only the personas relevant to the active profiles. |
+
+## Step 1 — read project state
+
+Read these files if they exist (they may not — that's information too):
+
+| Path | Tells you |
+|---|---|
+| `.wize/config/project.toml` | Active profiles, IDE targets, communication & document languages, project name. |
+| `.wize/config/tea.toml` | TEA gate policy (advisory vs enforcing). |
+| `.wize/planning/brief.md` | Whether Phase 1 (Pepper) started. |
+| `.wize/planning/research.md` | Whether research was done. |
+| `.wize/planning/ux/trigger-map.md` | Whether WDS trigger map exists. |
+| `.wize/planning/prd.md` | Whether Phase 2 (Maria Hill) PRD exists. |
+| `.wize/planning/ux/ux-scenarios.md`, `.wize/planning/ux/ux-design/**` | Whether Mantis ran UX. |
+| `.wize/planning/tech-vision.md`, `nfr-principles.md` | Whether Fury set tech strategy. |
+| `.wize/solutioning/architecture.md` | Whether Tony's architecture exists. |
+| `.wize/solutioning/stories/**/*.md` | Whether stories are sliced. |
+| `.wize/implementation/tea/risk-profile.md` | Whether Hawkeye's risk profile is in place. |
+| `.wize/implementation/tea/**/gate.md` | Last gate decisions per story. |
+| `.wize/implementation/sprint-status.md` | Active sprint state. |
+
+## Step 2 — determine current phase
+
+Apply this heuristic, top-down. Stop at the first match.
+
+1. **No `.wize/` folder.** → Tell the user the kit isn't installed; suggest `npx wize-dev-kit install`.
+2. **No `brief.md`.** → Phase 1. Next: **Pepper / `wize-product-brief`**.
+3. **`brief.md` exists, no `trigger-map.md`.** → Still Phase 1. Next: **Pepper / `wize-trigger-map`**.
+4. **No `prd.md`.** → Phase 2. Next: **Maria Hill / `wize-create-prd`**.
+5. **`prd.md` exists, no `ux-scenarios.md`.** → Phase 2 UX. Next: **Mantis / `wize-ux-scenarios`**.
+6. **`ux-design/` empty.** → Phase 2 UX continues. Next: **Mantis / `wize-ux-design`**.
+7. **No `tech-vision.md` or `nfr-principles.md`.** → Phase 2→3 boundary. Next: **Fury / `wize-tech-vision`** then `wize-nfr-principles`.
+8. **No `architecture.md`.** → Phase 3. Next: **Tony / `wize-create-architecture`**.
+9. **No `stories/**/*.md`.** → Phase 3 still. Next: **Tony / `wize-create-epics-and-stories`**.
+10. **No `tea/risk-profile.md`.** → Phase 3 closeout. Next: **Hawkeye / `wize-tea-risk`**.
+11. **Has stories but `sprint-status.md` shows no active sprint.** → Phase 4 start. Next: **Maria Hill / `wize-sprint-planning`**.
+12. **Has active sprint, oldest in-flight story has no `tea/.../design.md`.** → Next: **Hawkeye / `wize-tea-design`** for that story.
+13. **In-flight story exists, no implementation commits.** → Next: **Shuri / `wize-dev-story`** on that story.
+14. **In-flight story exists with code, no `gate.md`.** → Next: **Hawkeye / `wize-tea-trace` → `wize-tea-review` → `wize-tea-gate`** for that story.
+15. **All stories gated.** → Next: **Wizer / `wize-retrospective`** + plan next epic.
+
+For brownfield repos where `.wize/knowledge/document-project/` is missing, prepend: "Run `wize-document-project` first to baseline the codebase."
+
+## Step 3 — respond
+
+Default response shape (3 lines):
+
+```
+Welcome back. {project name} — {profiles, e.g., "Core + Web"}.
+You're at: {phase + last completed artifact}.
+Next: /{next workflow} ({persona}).
+```
+
+For `status`, return a markdown table:
+
+```
+| Item | State |
+|---|---|
+| Phase | … |
+| Profiles | … |
+| Last TEA gate | … (PASS/CONCERNS/FAIL/WAIVED) |
+| In-flight stories | … |
+| Active sprint | … |
+| TEA policy | advisory / enforcing |
+```
+
+For `personas`, list only personas whose role applies to active profiles. Always include Wizer, Pepper, Peggy, Maria Hill, Mantis, Fury, Tony, Hawkeye, Shuri (all are profile-independent core roles). If `web-overlay` active, note that **Mantis** has the WCAG/responsive playbook loaded and **Hawkeye** has Playwright/Vitest patterns. If `app-overlay` active, note HIG/Material 3 for Mantis and Detox/Maestro for Hawkeye.
+
+## Step 4 — offer to act
+
+End every response with one of:
+
+- "Want me to call {persona}?" — if the next step is clear.
+- "Want me to baseline the repo first?" — if brownfield + no document-project.
+- "Want me to call a party-mode with {persona1} + {persona2}?" — if the next step has a cross-cutting decision.
+
+## Style
+
+- Speak in the user's `communication` language (from `.wize/config/project.toml`).
+- One sharp question is better than three sentences of advice.
+- If the user invoked `/wize-help next`, give them just the next step in one line and stop.
+- If user invoked `/wize-help status`, give the table and stop. Don't suggest actions.

package/tools/installer/wize-cli.js

@@ -12,6 +12,9 @@
 const fs = require('node:fs');
 const path = require('node:path');
 const readline = require('node:readline');
+const prompts = require('prompts');
+
+const INTERACTIVE = process.stdout.isTTY && process.stdin.isTTY;
 
 const KIT_ROOT = path.resolve(__dirname, '..', '..');
 const KIT_VERSION = require(path.join(KIT_ROOT, 'package.json')).version;
@@ -34,6 +37,20 @@ const PROFILES = [
   { code: 'app-overlay', label: 'Wize App Development (overlay)', required: false }
 ];
 
+// Common BCP-47 short codes. Users can type any other value freely.
+const LANGUAGES = [
+  { code: 'en',    label: 'English' },
+  { code: 'pt-BR', label: 'Português (Brasil)' },
+  { code: 'pt-PT', label: 'Português (Portugal)' },
+  { code: 'es',    label: 'Español' },
+  { code: 'fr',    label: 'Français' },
+  { code: 'de',    label: 'Deutsch' },
+  { code: 'it',    label: 'Italiano' },
+  { code: 'zh-CN', label: '中文 (简体)' },
+  { code: 'ja',    label: '日本語' },
+  { code: 'vi',    label: 'Tiếng Việt' }
+];
+
 const HELP = `wize-dev-kit v${KIT_VERSION}
 
 Usage:
@@ -79,7 +96,69 @@ async function confirm(question, defaultYes = true) {
   return ans.startsWith('y');
 }
 
+async function selectLanguage(label, defaultCode = 'en') {
+  if (INTERACTIVE) {
+    const choices = LANGUAGES.map(l => ({
+      title: `${l.code.padEnd(6)} — ${l.label}`,
+      value: l.code,
+      description: l.code === defaultCode ? 'default' : undefined
+    }));
+    choices.push({ title: 'Other (type a custom BCP-47 code)…', value: '__custom' });
+    const initial = LANGUAGES.findIndex(l => l.code === defaultCode);
+    const { picked } = await prompts({
+      type: 'select',
+      name: 'picked',
+      message: label,
+      choices,
+      initial: initial === -1 ? 0 : initial
+    });
+    if (picked === undefined) process.exit(130);
+    if (picked === '__custom') {
+      const { custom } = await prompts({
+        type: 'text',
+        name: 'custom',
+        message: 'Type a BCP-47 language code (e.g. ko, ar, nl, ru)',
+        initial: defaultCode,
+        validate: v => /^[A-Za-z]{2,3}(-[A-Za-z0-9]{2,8})?$/.test(v) || 'use a valid code like "en" or "pt-BR"'
+      });
+      return custom || defaultCode;
+    }
+    return picked;
+  }
+  // Non-TTY fallback: number-driven.
+  console.log(`\n${label}:`);
+  LANGUAGES.forEach((l, i) => {
+    const def = l.code === defaultCode ? '(default)' : '';
+    console.log(`  ${String(i + 1).padStart(2)}. ${l.code.padEnd(6)} — ${l.label} ${def}`);
+  });
+  const ans = (await prompt(`Pick a number, type a code, or ENTER for ${defaultCode}: `)).trim();
+  if (!ans) return defaultCode;
+  const asNum = parseInt(ans, 10);
+  if (!isNaN(asNum) && asNum >= 1 && asNum <= LANGUAGES.length) return LANGUAGES[asNum - 1].code;
+  return ans;
+}
+
 async function multiSelect(label, items, isSelected = i => i.default) {
+  if (INTERACTIVE) {
+    const { picked } = await prompts({
+      type: 'multiselect',
+      name: 'picked',
+      message: label,
+      hint: '↑/↓ navigate · space toggle · enter confirm',
+      instructions: false,
+      choices: items.map(it => ({
+        title: it.label + (it.required ? ' (required)' : ''),
+        value: it.code,
+        selected: it.required || isSelected(it),
+        disabled: it.required
+      }))
+    });
+    if (picked === undefined) process.exit(130);
+    const codes = new Set(picked);
+    for (const req of items.filter(i => i.required)) codes.add(req.code);
+    return items.filter(it => codes.has(it.code));
+  }
+  // Non-TTY fallback.
   console.log(`\n${label}:`);
   items.forEach((it, i) => {
     const star = it.required ? '★ ' : '  ';
@@ -87,9 +166,7 @@ async function multiSelect(label, items, isSelected = i => i.default) {
     console.log(`  ${i + 1}. ${star}${it.label} ${def}`);
   });
   const ans = await prompt('Pick numbers (comma-separated) or ENTER for defaults: ');
-  if (!ans) {
-    return items.filter(it => it.required || isSelected(it));
-  }
+  if (!ans) return items.filter(it => it.required || isSelected(it));
   const indices = ans.split(',').map(s => parseInt(s.trim(), 10) - 1).filter(i => !isNaN(i));
   const picked = indices.map(i => items[i]).filter(Boolean);
   for (const req of items.filter(i => i.required)) {
@@ -147,7 +224,7 @@ const WIZE_DIRS = [
   '.wize/custom/workflows'
 ];
 
-function projectToml({ profiles, targets, language, project_name }) {
+function projectToml({ profiles, targets, communication_language, document_output_language, project_name }) {
   const profileLine = profiles.map(p => `"${p.code}"`).join(', ');
   const targetLine = targets.map(t => `"${t.code}"`).join(', ');
   return `# Wize Development Kit — project config
@@ -162,8 +239,11 @@ profiles = [${profileLine}]
 ide_targets = [${targetLine}]
 
 [language]
-communication = "${language}"
-document_output = "${language}"
+# Language used when agents talk to you in the IDE chat.
+communication = "${communication_language}"
+# Language used when agents write artifacts to disk
+# (brief.md, prd.md, architecture.md, gate.md, etc.).
+document_output = "${document_output_language}"
 
 [paths]
 output_root    = ".wize"
@@ -197,6 +277,35 @@ function userToml() {
 `;
 }
 
+function renderAdapters({ kitRoot, projectRoot, targets, profiles }) {
+  const results = [];
+  const profileCodes = profiles.map(p => p.code);
+  for (const target of targets) {
+    const adapterDir = path.join(kitRoot, 'adapters', target.code);
+    const renderPath = path.join(adapterDir, 'render.js');
+    if (!fs.existsSync(renderPath)) {
+      results.push({ code: target.code, skipped: true, reason: 'adapter missing' });
+      continue;
+    }
+    try {
+      const mod = require(renderPath);
+      if (typeof mod.render !== 'function') {
+        results.push({ code: target.code, skipped: true, reason: 'no render() export' });
+        continue;
+      }
+      const out = mod.render(kitRoot, projectRoot, { profiles: profileCodes });
+      if (out && typeof out === 'object' && Array.isArray(out.written)) {
+        results.push({ code: target.code, written: out.written.length });
+      } else {
+        results.push({ code: target.code, skipped: true, reason: 'stub (no skills emitted)' });
+      }
+    } catch (err) {
+      results.push({ code: target.code, error: err.message });
+    }
+  }
+  return results;
+}
+
 async function cmdInstall(args) {
   const cwd = process.cwd();
   console.log(logo());
@@ -217,12 +326,22 @@ async function cmdInstall(args) {
   const project_name = (await prompt(`Project name [${path.basename(cwd)}]: `)) || path.basename(cwd);
   const profiles = await multiSelect('Select profile(s) to install', PROFILES);
   const targets = await multiSelect('Select IDE target(s)', TARGETS);
-  const language = (await prompt('Communication language [en]: ')) || 'en';
+
+  const communication_language = await selectLanguage(
+    'Communication language (how agents will talk to you in chat)',
+    'en'
+  );
+  const document_output_language = await selectLanguage(
+    `Document output language (language used in generated files; ENTER for "${communication_language}")`,
+    communication_language
+  );
 
   console.log('\nCreating .wize/ skeleton...');
   for (const dir of WIZE_DIRS) mkdirp(path.join(cwd, dir));
 
-  writeIfMissing(path.join(cwd, '.wize/config/project.toml'), projectToml({ profiles, targets, language, project_name }));
+  writeIfMissing(path.join(cwd, '.wize/config/project.toml'), projectToml({
+    profiles, targets, communication_language, document_output_language, project_name
+  }));
   writeIfMissing(path.join(cwd, '.wize/config/user.toml'), userToml());
   writeIfMissing(path.join(cwd, '.wize/config/tea.toml'), teaToml());
 
@@ -230,8 +349,18 @@ async function cmdInstall(args) {
   console.log(`✓ profiles: ${profiles.map(p => p.code).join(', ')}`);
   console.log(`✓ ide targets: ${targets.map(t => t.code).join(', ')}`);
 
-  console.log('\n(stub) IDE adapter generation would run now for each target.');
-  console.log('(stub) onboarding handoff to Wizer would start here.');
+  console.log('\nGenerating IDE adapters...');
+  const renderResults = renderAdapters({
+    kitRoot: KIT_ROOT,
+    projectRoot: cwd,
+    targets,
+    profiles
+  });
+  for (const r of renderResults) {
+    if (r.skipped) console.log(`  - ${r.code}: skipped (${r.reason})`);
+    else if (r.written) console.log(`  ✓ ${r.code}: ${r.written} skill(s) emitted`);
+    else if (r.error) console.log(`  ✖ ${r.code}: ${r.error}`);
+  }
 
   if (detection.brownfield) {
     const baseline = await confirm('\nRun `wize-document-project` to baseline the existing repo now?', true);

package/src/core-skills/wize-help/skill.md

@@ -1,18 +0,0 @@
----
-code: wize-help
-name: Help
-module: core
-status: stub
----
-
-# Help
-
-Routes user intent to the right agent or workflow.
-
-## Examples
-- "I need to start a new project." → Wizer onboarding.
-- "I have a brief; what's next?" → Maria Hill / `wize-create-prd`.
-- "Design system?" → Mantis / `wize-design-system`.
-- "Architecture?" → Tony / `wize-create-architecture`.
-- "Test plan for this story?" → Hawkeye / `wize-tea-design`.
-- "Add a custom agent." → Builder / `wize-create-agent`.