Requirements
- Target platform
- OpenClaw
- Install method
- Manual import
- Extraction
- Extract archive
- Prerequisites
- OpenClaw
- Primary doc
- SKILL.md
Tencent SkillHub ยท Productivity
Parallel Agents
Spawns real AI-powered OpenClaw sub-sessions to run multiple specialized agents concurrently for content, dev, QA, docs, and autonomous workflows.
skill openclawclawhub Free
Spawns real AI-powered OpenClaw sub-sessions to run multiple specialized agents concurrently for content, dev, QA, docs, and autonomous workflows.
โฌ 0 downloads โ
0 stars Unverified but indexed
Install for OpenClaw
Quick setup
- Download the package from Yavira.
- Extract the archive and review SKILL.md first.
- Import or place the package into your OpenClaw setup.
Package facts
- Download mode
- Yavira redirect
- Package format
- ZIP package
- Source platform
- Tencent SkillHub
- What's included
- README.md, SKILL.md, USAGE-GUIDE.md, ai_orchestrator.py, examples/README.md, examples/simple_parallel_research.py
Validation
- Use the Yavira download entry.
- Review SKILL.md after the package is downloaded.
- Confirm the extracted package contains the expected setup assets.
Install with your agent
Agent handoff
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
New install
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete.
Upgrade existing
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run.
Trust & source
Release facts
- Source
- Tencent SkillHub
- Verification
- Indexed source record
- Version
- 3.2.0
Provenance
- Publisher
- jdalbright
- Source page
- View original listing
- Canonical URL
- Open canonical page
Documentation
Parallel Agents Skill - REAL AI Edition
๐ Execute tasks with ACTUAL AI-powered parallel agents using OpenClaw's sessions_spawn. โ ๏ธ HONEST STATUS: This skill has been rewritten to use REAL AI via sessions_spawn. Previously it simulated agents with templates. Now it ACTUALLY spawns AI sub-sessions.
๐จ CRITICAL USAGE NOTE
The orchestrator MUST be called from within an OpenClaw agent session, NOT as a standalone script. Why? The tools module (which provides sessions_spawn) is only available in the agent's runtime context, not in subprocess/exec calls. โ CORRECT: Call sessions_spawn directly from agent code (see USAGE-GUIDE.md) โ INCORRECT: Run orchestrator as standalone Python script via exec/subprocess ๐ SEE: USAGE-GUIDE.md for tested working examples and patterns
๐ฏ Capabilities
This skill provides 4 levels of agent automation: LevelFeatureWhat It Does1Task Agents (16 types)Specialized agents for content, dev, QA, docs2Meta Agents (4 types)Agents that create, review, refine, and orchestrate other agents3Iterative RefinementAutomatic quality improvement loop (Creator โ Reviewer โ Refiner)4Agent OrchestratorFully autonomous workflow management - just ask and it handles everything Proven Capabilities: โ 20 concurrent agents spawned simultaneously โ Smart model hierarchy - Haiku โ Kimi โ Opus (cost optimization) โ Auto-escalation - Agents automatically use better models if needed โ 100% success rate on mass creation tests with hierarchy โ 3/3 agents refined to 8.5+ quality in single iteration โ 4-agent hierarchy for complete autonomy
What This Actually Does
This skill creates real AI sub-sessions using OpenClaw's sessions_spawn tool. Each "agent" is: A spawned OpenClaw session (not a subprocess) Running real AI (same model as the host) Completely isolated from other agents Able to use all the same tools as the host Previous version: Subprocess workers with templates โ Current version: Real spawned AI sessions โ
Requirements
Must be run inside an OpenClaw session (for sessions_spawn access) OpenClaw gateway must be running The sessions tool must be available in your environment
โ Correct Usage: Direct sessions_spawn Calls
From within an OpenClaw agent (like Scout): # Spawn multiple agents in parallel using sessions_spawn tool directly from tools import sessions_spawn # Agent 1: Research task result1 = sessions_spawn( task="Research and provide: Top 3 gay-friendly bars in Savannah. Return as JSON.", runTimeoutSeconds=90, cleanup="delete" ) # Agent 2: Different research task result2 = sessions_spawn( task="Research and provide: Best restaurants for birthday dinner. Return as JSON.", runTimeoutSeconds=90, cleanup="delete" ) # Agent 3: Another parallel task result3 = sessions_spawn( task="Research and provide: Top photo spots in Savannah. Return as JSON.", runTimeoutSeconds=90, cleanup="delete" ) # All 3 agents now running in parallel! # Check results with sessions_list() and sessions_history()
โ Incorrect Usage: Standalone Script
# This WON'T work - tools module not available in subprocess python3 ~/.openclaw/skills/parallel-agents/ai_orchestrator.py
Basic Usage
from ai_orchestrator import RealAIParallelOrchestrator, AgentTask # Create orchestrator orch = RealAIParallelOrchestrator(max_concurrent=5) # Define tasks tasks = [ AgentTask( agent_type='content_writer_funny', task_description='Write a caption about gym life', input_data={'tone': 'motivational'} ), AgentTask( agent_type='content_writer_creative', task_description='Write a caption about gym life', input_data={'tone': 'inspirational'} ), ] # Execute in parallel (ACTUALLY spawns AI sessions) results = orch.run_parallel(tasks)
How It Works
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ Main Session โ โ (Your OpenClaw Instance) โ โ ๐ง Host AI โ โโโโโโโโโโโโโโโโโโโโโโโฌโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ sessions_spawn (REAL) โ โโโโโโโโโโโโโโโผโโโโโโโโโโโโโโฌโโโโโโโโโโโโโโ โ โ โ โ โโโโโโผโโโโโ โโโโโโผโโโโโ โโโโโโผโโโโโ โโโโโโผโโโโโ โ Agent 1 โ โ Agent 2 โ โ Agent 3 โ โ Agent N โ โ ๐ โ โ ๐ป โ โ ๐ โ โ ๐จ โ โ REAL AI โ โ REAL AI โ โ REAL AI โ โ REAL AI โ โ Session โ โ Session โ โ Session โ โ Session โ โโโโโโโโโโโ โโโโโโโโโโโ โโโโโโโโโโโ โโโโโโโโโโโ
The sessions_spawn Integration
Each agent is spawned with: from tools import sessions_spawn result = sessions_spawn( task=agent_prompt, # Full task description agent_id=f"agent_{type}_{id}", # Unique identifier model="kimi-coding/k2p5", # AI model runTimeoutSeconds=120, # Max execution time cleanup="delete" # Auto-cleanup )
Content Writers
Agent TypePurposeSystem Promptcontent_writer_creativeImaginative, artisticRich metaphors, emotional resonancecontent_writer_funnyHumorous, wittyJokes, wordplay, relatable humorcontent_writer_educationalTeaching contentClear explanations, actionable takeawayscontent_writer_trendyViral contentTrend-aware, culturally relevantcontent_writer_controversialDebate-sparkingHot takes, respectful discourse
Development Agents
Agent TypePurposeOutputfrontend_developerReact/Vue/AngularComponent structure, state managementbackend_developerFastAPI/Flask/DjangoAPI endpoints, auth, modelsdatabase_architectSchema designTables, indexes, migrationsapi_designerREST/GraphQLOpenAPI specs, rate limitsdevops_engineerCI/CDDocker, K8s, pipelines
QA Agents
Agent TypePurposeFocuscode_reviewerQuality reviewBest practices, maintainabilitysecurity_reviewerSecurity scanVulnerabilities, threatsperformance_reviewerOptimizationBottlenecks, complexityaccessibility_reviewerWCAG complianceA11y, screen readerstest_engineerTest coverageUnit/integration tests
Documentation
Agent TypePurposedocumentation_writerREADMEs, API docs, guides
Personalized Agents (Jake's Suite) ๐พ
Agents created specifically for Jake's needs via agent_orchestrator research: Agent TypePurposeKey Featurestravel_event_plannerTrip content coordinationSavannah/Atlanta/SD Pride planning, gear checklists, event schedulesdonut_care_coordinatorPrincess Donut managementFeeding tracking, vet reminders, pet sitter coordination, daily updatespup_community_engagerPup community managementBluesky/Twitter monitoring, DM triage, authentic pup voice engagementprint_project_manager3D printing workflowModel queue, filament tracking, vibecoding integration, print optimizationtraining_assistantAlmac work productivityTraining prep, onboarding, session checklists, material templates Total Agent Types: 25 5 Content Writers 5 Development Agents 5 QA Agents 1 Documentation Agent 5 Personalized Agents ๐ 4 Meta Agents
Meta Agents ๐ (Agent Creation System)
Agent TypePurposeWhat It Doesagent_creatorDesigns new AI agentsCreates complete agent definitions with prompts, schemas, examplesagent_design_reviewerValidates agent designsReviews quality, completeness, production readiness (scores 0-10)agent_refinerImproves agent designsApplies fixes based on review feedback to reach target scoresagent_orchestratorMaster coordinatorPlans workflows, spawns agents, coordinates execution, compiles results The 4-Agent Hierarchy: Level 4: USER โ asks Level 3: AGENT_ORCHESTRATOR โ plans, spawns, coordinates Level 2: Meta Agents (creator, reviewer, refiner) โ designs, reviews, refines Level 1: Task Agents (content writers, developers, QA) โ does work Level 0: Actual Tasks Total Agent Types: 20 5 Content Writers 5 Development Agents 5 QA Agents 1 Documentation Agent 4 Meta Agents ๐ Workflow 1: Simple Creation (2 agents) from ai_orchestrator import ( RealAIParallelOrchestrator, create_meta_agent_workflow ) orch = RealAIParallelOrchestrator() # Define agents to create new_agents = [ {'name': 'crypto_analyst', 'purpose': 'Analyze crypto trends'}, {'name': 'content_strategist', 'purpose': 'Plan content calendars'} ] # Creates: 2 creators + 2 reviewers (4 tasks) tasks = create_meta_agent_workflow(new_agents) results = orch.run_parallel(tasks) Workflow 2: Iterative Refinement (3-agent loop) # The full 3-agent refinement workflow: # Creator โ Reviewer (scores) โ Refiner (fixes) โ Reviewer (verifies) # Repeats until score >= 8.5 agents_to_refine = [ {'name': 'my_agent', 'current_score': 7.4, 'target': 8.5} ] # This runs the full loop automatically results = orch.run_iterative_refinement(agents_to_refine) # Result: 7.4 โ 8.5+ โ Workflow 3: Orchestrated Mass Creation (autonomous) # Spawn the orchestrator to handle everything: # - Plans workflow # - Spawns all agents # - Coordinates execution # - Handles refinements # - Compiles final report result = sessions_spawn( task="Create 5 new agents and ensure all score 8.5+", agent_type='agent_orchestrator', timeout=600 ) # The orchestrator does everything autonomously! This enables agent bootstrapping - the system creates and improves itself!
AgentTask
@dataclass class AgentTask: agent_type: str # Type from registry (required) task_description: str # What to do (required) input_data: Dict # Input parameters (optional) task_id: str # Unique ID (auto-generated) timeout_seconds: int # Max time (default: 120) output_format: str # json|markdown|code|text
AgentResult
@dataclass class AgentResult: task_id: str # Matches AgentTask agent_type: str # Agent that produced this status: str # pending|running|completed|failed output: Any # Generated content (agent-dependent format) execution_time: float # Time taken error: str # Error message if failed session_key: str # Spawned session identifier
Example 1: Generate Multiple Content Styles
from ai_orchestrator import RealAIParallelOrchestrator, create_content_team orch = RealAIParallelOrchestrator(max_concurrent=5) tasks = create_content_team("Monday motivation", platform="bluesky") # This spawns 5 REAL AI agents results = orch.run_parallel(tasks) print("Agents spawned! Each is generating content...") print("Check sessions_list() to see running agents")
Example 2: Full-Stack Development Team
from ai_orchestrator import RealAIParallelOrchestrator, create_dev_team orch = RealAIParallelOrchestrator(max_concurrent=5) tasks = create_dev_team("TaskManager", ['auth', 'tasks', 'teams']) # Spawns 5 dev agents in parallel results = orch.run_parallel(tasks) # Each agent designs their layer independently # - Frontend agent designs React components # - Backend agent designs FastAPI routes # - Database agent designs schema # - etc.
Example 3: Code Review Team
from ai_orchestrator import RealAIParallelOrchestrator, create_review_team code = open('app.py').read() orch = RealAIParallelOrchestrator(max_concurrent=5) tasks = create_review_team(code) # Spawns 5 reviewers simultaneously results = orch.run_parallel(tasks) # Each reviews from different angle: # - Code quality # - Security # - Performance # - Accessibility # - Test coverage
Example 4: Meta-Agent System (Agents Creating Agents) ๐
from ai_orchestrator import ( RealAIParallelOrchestrator, create_meta_agent_workflow ) orch = RealAIParallelOrchestrator(max_concurrent=6) # Define new agents to create new_agents = [ { 'name': 'social_media_analyst', 'purpose': 'Analyze social media performance', 'domain': 'social media analytics', 'capabilities': ['engagement analysis', 'trend identification'] }, { 'name': 'bug_hunter', 'purpose': 'Find bugs in code', 'domain': 'software QA', 'capabilities': ['static analysis', 'edge case detection'] }, { 'name': 'api_documenter', 'purpose': 'Generate API docs', 'domain': 'technical writing', 'capabilities': ['endpoint extraction', 'example generation'] } ] # Creates 6 tasks: 3 creators + 3 reviewers tasks = create_meta_agent_workflow(new_agents) results = orch.run_parallel(tasks) # Result: 3 complete agent definitions + 3 quality reviews # All created entirely by AI in parallel! This is agent bootstrapping - the system creates itself!
Example 5: Mass Agent Creation (10+ Agents at Once) ๐ฅ
Proven Capability: The system has been tested with 20 concurrent agents (10 creators + 10 reviewers) all spawned simultaneously. from ai_orchestrator import RealAIParallelOrchestrator, AgentTask orch = RealAIParallelOrchestrator(max_concurrent=10) # Define 10 new agents to create new_agents = [ {'name': 'engagement_optimizer', 'purpose': 'Analyze social media posts', 'domain': 'social media', 'capabilities': ['analytics', 'optimization']}, {'name': 'workout_designer', 'purpose': 'Create gym/home workouts', 'domain': 'fitness', 'capabilities': ['program design', 'adaptation']}, {'name': 'email_drafter', 'purpose': 'Write professional/personal emails', 'domain': 'communication', 'capabilities': ['tone adaptation', 'drafting']}, # ... more agents ] # Create all 10 agents + 10 reviewers = 20 parallel agents! all_tasks = [] for agent in new_agents: # Add creator all_tasks.append(AgentTask( agent_type='agent_creator', task_description=f"Design agent: {agent['name']}", input_data=agent, timeout_seconds=180 )) # Add reviewer all_tasks.append(AgentTask( agent_type='agent_design_reviewer', task_description=f"Review {agent['name']}", input_data={'agent_name': agent['name']}, timeout_seconds=120 )) # SPAWN 20 AGENTS SIMULTANEOUSLY results = orch.run_parallel(all_tasks) Real-World Results (2026-02-08 Test): โ 10 Agent Creators spawned successfully โ 10 Design Reviewers spawned successfully โ All 20 completed without errors โ Average quality score: 8.1/10 โ Production-ready agent definitions created Practical Limit: ~20-50 concurrent agents (depends on system resources) See: examples/mass_agent_creation.py for full implementation.
Collecting Results
Agents return their output in their session transcript. To collect: # After spawning, poll for results from tools import sessions_list, sessions_history # Check which agents have completed sessions = sessions_list(agent_id_pattern="agent_*") for session in sessions: if session['status'] == 'completed': history = sessions_history(session['sessionKey']) # Parse JSON from final assistant message output = json.loads(history[-1]['content']) Note: Full result collection is implemented in the orchestrator. Results are available via results attribute after spawning.
Why sessions_spawn?
Previous implementations tried: Threading - Limited by Python GIL, not truly parallel Multiprocessing - macOS spawn issues, complex IPC Subprocess workers - Templates, not real AI sessions_spawn is the solution: True isolation (separate sessions) Full AI capabilities (same model) Built into OpenClaw Automatic cleanup
Limitations
OpenClaw dependency - Must run inside OpenClaw session Result collection - Requires polling sessions_list Cost - Each spawn = separate API call (but same model/credentials) Timeout - Agents limited to 120 seconds by default
File Structure
~/.openclaw/skills/parallel-agents/ โโโ README.md # Quick start guide โโโ SKILL.md # Complete documentation โโโ USAGE-GUIDE.md # Practical examples and patterns โโโ ai_orchestrator.py # Core orchestrator code โโโ helpers.py # Auto-retry helper functions โโโ examples/ # Working examples โโโ README.md # Examples documentation โโโ simple_parallel_research.py # Simple example
Version History
3.2.0 (2026-02-08): SMART MODEL HIERARCHY โ Added intelligent model escalation (Haiku โ Kimi โ Opus) โ Cost optimization: Try cheapest model first, escalate if needed โ Updated helpers.py with spawn_with_model_hierarchy() โ Auto-escalation in spawn_with_retry() and spawn_parallel_with_retry() โ Comprehensive docs on model selection and cost savings โ Tested: Haiku completes simple tasks successfully 3.1.0 (2026-02-08): PRODUCTION READY โ Added auto-retry helpers (spawn_with_retry, spawn_parallel_with_retry) โ Cleaned up development artifacts (removed 18 outdated files) โ Added comprehensive documentation (README, USAGE-GUIDE) โ Simplified examples (one clear working example) โ Tested in production (Savannah trip research) โ Published to ClawHub 3.0.0 (2026-02-08): NUCLEAR OPTION - REAL AI AGENTS Complete rewrite to use sessions_spawn Each agent is a real spawned AI session No more simulation or templates Requires OpenClaw environment
"sessions_spawn not available"
Cause: Not running inside OpenClaw session Fix: Run your script inside OpenClaw
"No module named 'tools'"
Cause: Outside OpenClaw environment Fix: The sessions tool is only available inside OpenClaw
Agents fail immediately
Cause: OpenClaw gateway not running Fix: Start gateway: openclaw gateway start
This Actually Spawns Real AI Now
No more simulation. No more templates. When you run this inside OpenClaw: Real sessions_spawn calls happen Real AI sub-sessions are created Real reasoning occurs in each agent Real JSON output is generated The agents don't just execute code โ they think, create, and analyze independently using genuine AI cognition. Welcome to actual parallel AI. ๐ Built for OpenClaw using real sessions_spawn technology. Part of the OpenClaw skill ecosystem. Honest Edition: No simulation, just real AI.
Category context
Workflow acceleration for inboxes, docs, calendars, planning, and execution loops.
Source: Tencent SkillHub
Largest current source with strong distribution and engagement signals.
Package contents
Included in package
4 Docs2 Scripts
- SKILL.md
- examples/README.md
- README.md
- USAGE-GUIDE.md
- ai_orchestrator.py
- examples/simple_parallel_research.py