cc-workspace 4.0.0

package/global-skills/rules/model-routing.md

@@ -0,0 +1,42 @@
+---
+description: Model routing rules. The orchestrator never writes in repos — it delegates to teammates. Can write in orchestrator/.
+globs: ["workspace.md", "plans/**", "constitution.md", "templates/**"]
+---
+
+# Model Routing
+
+## Golden rule
+The orchestrator NEVER writes code in repos. It describes what teammates must do.
+Writing plans, workspace.md, constitution.md in orchestrator/ is expected and normal.
+If you write code for a repo (not a markdown plan), you have failed — delegate.
+
+## Security layers (3 layers in agent frontmatter — no settings.json pollution)
+1. **Agent frontmatter**: `disallowedTools` — refused at model level
+2. **Agent frontmatter**: `allowed-tools` — whitelist of permitted tools
+3. **Agent hook**: `PreToolUse` in frontmatter — structured deny response
+
+## Routing table
+| Role | Model | Mechanism |
+|------|-------|-----------|
+| Orchestrator | **Opus 4.6** | `claude --agent team-lead` (frontmatter `model: opus`) |
+| Implementation teammates | **Sonnet 4.6** | `CLAUDE_CODE_SUBAGENT_MODEL=sonnet` |
+| QA investigators | **Sonnet 4.6** | Same |
+| Explorers / cross-checks | **Haiku** | `model: haiku` in skill/agent frontmatter |
+| Plan review | **Haiku** | `model: haiku` in skill frontmatter |
+
+## Effort levels
+| Role | Effort | Rationale |
+|------|--------|-----------|
+| Orchestrator (team-lead) | **high** | Complex planning, multi-step reasoning |
+| Implementation teammates | **medium** | Focused implementation, clear scope |
+| QA investigators | **medium** | Thorough but bounded analysis |
+| Explorers / cross-checks (Haiku) | **low** | Fast scans, read-only |
+| Plan review (Haiku) | **low** | Structural checks, no deep reasoning |
+
+Effort levels are set in agent frontmatter (`effort: high|medium|low|max`).
+Use `max` only for critical debugging or complex architectural decisions.
+
+## Custom agent `implementer`
+For Task subagents that need to write code in an isolated worktree,
+use `@implementer` (frontmatter with `isolation: worktree`).
+Agent Teams teammates (Teammate tool) get automatic isolation.

package/global-skills/templates/claude-md.template.md

@@ -0,0 +1,124 @@
+# CLAUDE.md — [REPO_NAME]
+
+> [REPO_DESCRIPTION — one line]
+
+## Stack technique
+
+| Technologie | Version | Notes |
+|-------------|---------|-------|
+| [language] | [version] | |
+| [framework] | [version] | |
+| [database] | [version] | [usage] |
+| [test framework] | [version] | |
+
+## Architecture
+
+<!-- Describe the architecture pattern and show the directory tree -->
+
+```
+[repo]/
+├── src/           ← [description]
+├── tests/         ← [description]
+└── ...
+```
+
+<!-- Explain the layering or module organization -->
+
+## Regles critiques
+
+<!-- Numbered rules. Each rule has a short name and ✅/❌ examples.
+     Keep only rules that an AI agent MUST know to avoid mistakes.
+     Not a style guide — only rules where violations cause bugs or break patterns. -->
+
+### 1. [Rule name]
+
+[One-line explanation]
+
+```
+❌ [wrong pattern]
+✅ [correct pattern]
+```
+
+### 2. [Rule name]
+
+[One-line explanation]
+
+```
+❌ [wrong pattern]
+✅ [correct pattern]
+```
+
+<!-- Add more rules as needed. Typical count: 3-8 rules per repo. -->
+
+## Conventions de nommage
+
+| Type | Pattern | Exemple |
+|------|---------|---------|
+| [files] | [pattern] | [example] |
+| [classes] | [pattern] | [example] |
+| [functions] | [pattern] | [example] |
+| [routes/endpoints] | [pattern] | [example] |
+
+## Tests
+
+<!-- Test framework, how to run, patterns to follow -->
+
+**Lancer les tests :**
+```bash
+[test command]
+```
+
+**Patterns :**
+- [describe test organization]
+- [describe mocking strategy]
+- [minimum coverage if applicable]
+
+## Anti-patterns
+
+<!-- Two-column table: forbidden vs required. Keep it short and impactful. -->
+
+| Interdit | Faire plutot |
+|----------|--------------|
+| [bad pattern] | [good pattern] |
+| [bad pattern] | [good pattern] |
+| [bad pattern] | [good pattern] |
+
+## Commandes
+
+<!-- Essential commands for dev workflow -->
+
+| Commande | Description |
+|----------|-------------|
+| `[cmd]` | Build / compile |
+| `[cmd]` | Run tests |
+| `[cmd]` | Lint / format |
+| `[cmd]` | Start dev server |
+
+## Domaines existants
+
+<!-- List existing modules/domains/packages if the repo uses a modular architecture.
+     Remove this section if not applicable. -->
+
+| Domaine | Chemin | Description |
+|---------|--------|-------------|
+| [name] | [path] | [one-line description] |
+
+<!-- === OPTIONAL SECTIONS (add only if relevant) === -->
+
+<!-- ## Multi-tenancy
+     Describe tenant scoping pattern, trait/mixin used, tables affected. -->
+
+<!-- ## Auth
+     Describe authentication mechanism, tokens, middleware, guards. -->
+
+<!-- ## Checklist avant livraison
+     - [ ] Tests passent
+     - [ ] Lint clean
+     - [ ] Pas de code mort introduit
+     - [ ] Conventions respectees -->
+
+<!-- ## Decisions architecturales
+     Document any ADRs (Architecture Decision Records) specific to this repo. -->
+
+---
+_Last Updated: [DATE]_

package/global-skills/templates/constitution.template.md

@@ -0,0 +1,18 @@
+# Constitution — [PROJECT_NAME]
+> Configured on: [DATE]
+
+> The 12 universal principles (security, UX, code, process) apply automatically.
+> This file contains ONLY project-specific rules, numbered from 13.
+
+## Project-specific rules
+
+13. **[Rule name].** [Description — what, why, and when it applies]
+    <!-- Example: **Data precision.** Financial amounts stored as integers (cents). Never floats. -->
+
+14. **[Rule name].** [Description]
+    <!-- Example: **Multi-tenancy.** Every query, endpoint, and view scoped by tenant_id. No exceptions. -->
+
+15. **[Rule name].** [Description]
+    <!-- Example: **Irreversible operations.** Any irreversible DB change requires feature flag. -->
+
+<!-- Add more rules as needed. Keep each rule actionable and verifiable. -->

package/global-skills/templates/workspace.template.md

@@ -0,0 +1,33 @@
+# Workspace: [PROJECT_NAME]
+> Configured on: [DATE]
+> Orchestrator v4.0
+
+## Projet
+<!-- What does this project do? Who is it for? Scale? -->
+[UNCONFIGURED]
+
+## Service map
+<!-- Auto-generated from sibling repo scan. Review and add descriptions. -->
+| Service | Repo | Type | CLAUDE.md | Description |
+|---------|------|------|-----------|-------------|
+<!-- [AUTO_DISCOVERED_REPOS] -->
+
+## Inter-service relationships
+<!-- Describe the request flow between services -->
+```
+[service-a] → [service-b] → [database]
+```
+
+## Business rules
+<!-- Key rules that teammates must know. Examples:
+- Amounts in cents (no floats)
+- Multi-tenant: all queries scoped by company_id
+- Audit logs mandatory for all mutations
+-->
+
+## Onboarding notes
+<!-- Anything the orchestrator should tell teammates that isn't in CLAUDE.md files.
+- Common pitfalls
+- Local setup requirements
+- Non-obvious dependencies
+-->

package/package.json

@@ -0,0 +1,28 @@
+{
+  "name": "cc-workspace",
+  "version": "4.0.0",
+  "description": "Claude Code multi-workspace orchestrator — skills, hooks, agents, and templates for multi-service projects",
+  "bin": {
+    "cc-workspace": "./bin/cli.js"
+  },
+  "files": [
+    "bin/",
+    "global-skills/",
+    "LICENSE",
+    "README.md"
+  ],
+  "keywords": [
+    "claude",
+    "claude-code",
+    "orchestrator",
+    "agent-teams",
+    "multi-repo",
+    "workspace",
+    "monorepo"
+  ],
+  "author": "Vincent",
+  "license": "MIT",
+  "engines": {
+    "node": ">=18"
+  }
+}