Orchestrate parallel development with multiple Claude Code agents from PRD specs. Use when asked to parallelize development, break down a PRD into agent tasks, coordinate multi-agent workflows, or scale development across independent workstreams.
Orchestrates parallel development by decomposing PRDs into independent tasks for multiple Claude agents. Use when asked to parallelize work, break down specifications, or coordinate multi-agent workflows across isolated git worktrees.
/plugin marketplace add jpoutrin/product-forge/plugin install product-design@product-forge-marketplaceThis skill inherits all available tools. When active, it can use any tool Claude has access to.
Orchestrate massively parallel development by decomposing work into independent tasks that multiple Claude Code instances can execute simultaneously.
The cpo CLI tool handles parallel agent execution with git worktree isolation.
pip install claude-parallel-orchestrator
# or
pipx install claude-parallel-orchestrator
| Command | Description |
|---|---|
cpo init <dir> -t <tech-spec> -n <name> | Initialize parallel directory |
cpo validate <dir> | Validate manifest and structure |
cpo run <dir> | Execute parallel agents |
cpo status <dir> | Check execution status |
/parallel-setup -> One-time: creates parallel/ directory
|
/parallel-decompose -> Per Tech Spec: creates TS-XXXX-slug/ with all artifacts
|
/parallel-run -> Delegates to `cpo run` for execution
|
/parallel-integrate -> Verify & generate integration report
Each decomposition creates an isolated artifact folder keyed by Tech Spec:
project/
parallel/ # Created by /parallel-setup (one-time)
README.md
.gitignore
TS-XXXX-{slug}/ # Created by /parallel-decompose
manifest.json # Regeneration metadata
context.md # Shared project context (token-efficient)
architecture.md # System design from Tech Spec
task-graph.md # Dependency visualization (Mermaid)
contracts/
types.py (or types.ts) # Shared domain types
api-schema.yaml # OpenAPI specification
tasks/
task-001-users.md # Compact YAML format
task-002-products.md
...
prompts/
agent-prompts.md # All launch commands
task-*.txt # Individual agent prompts
integration-report.md # Post-execution report
tech-specs/ # Source Tech Specs
approved/TS-XXXX-slug.md
CLAUDE.md # Project conventions
This skill is part of a family of parallel development skills:
| Skill | Purpose |
|---|---|
| parallel-decompose | PRD decomposition workflow, task generation, contracts |
| parallel-prompt-generator | Generate agent prompts from task specs |
| parallel-execution | Git worktrees, parallel execution patterns, scripts |
| parallel-task-format | Task spec YAML format, scope notation, agent selection |
| agent-tools | Tool permissions, CLI syntax for agent restrictions |
/parallel-setup --tech django
Creates parallel/ directory structure.
/parallel-decompose docs/prd.md --tech-spec tech-specs/approved/TS-0042-inventory.md
Creates parallel/TS-0042-inventory-system/ with:
# Using /parallel-run (delegates to cpo)
/parallel-run parallel/TS-0042-inventory-system/
# Or using cpo directly
cpo run parallel/TS-0042-inventory-system/
/parallel-integrate --parallel-dir parallel/TS-0042-inventory-system
Checks contract compliance, boundary compliance, runs tests, generates report.
{
"tech_spec_id": "TS-0042",
"name": "inventory-system",
"technology": "python",
"python_version": "3.11",
"waves": [
{
"number": 1,
"tasks": [
{ "id": "task-001", "agent": "python-experts:django-expert" },
{ "id": "task-002", "agent": "python-experts:django-expert" }
],
"validation": "from apps.users.models import User; print('Wave 1 OK')"
},
{
"number": 2,
"tasks": [
{ "id": "task-003", "agent": "python-experts:django-expert" }
],
"validation": "from apps.orders.models import Order; print('Wave 2 OK')"
}
],
"metadata": {
"tech_spec": "tech-specs/approved/TS-0042-inventory.md",
"generated_at": "2025-01-15T10:00:00Z",
"total_tasks": 3,
"max_parallel": 2,
"critical_path": ["task-001", "task-003"]
}
}
| Command | Purpose |
|---|---|
/parallel-setup | One-time project initialization |
/parallel-decompose | Per-spec decomposition with prompts |
/parallel-run | Execute and monitor parallel agents |
/parallel-integrate | Post-execution verification |
/create-tech-spec | Create Tech Spec before decomposition |
This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.