Requirements
- Target platform
- OpenClaw
- Install method
- Manual import
- Extraction
- Extract archive
- Prerequisites
- OpenClaw
- Primary doc
- SKILL.md
Coordinate multiple agents and tasks for complex workflows. Orchestrate subagents, manage dependencies, handle parallel execution, and ensure successful comp...
Coordinate multiple agents and tasks for complex workflows. Orchestrate subagents, manage dependencies, handle parallel execution, and ensure successful comp...
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
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. Summarize what changed and any follow-up checks I should run.
Coordinate multiple agents and tasks for complex workflows.
Multi-step operations requiring coordination Parallel execution of independent tasks Complex workflows with dependencies Orchestrating subagents for large projects
Break down complex tasks into manageable steps Manage dependencies between tasks Coordinate parallel execution Handle task sequencing and scheduling
Spawn and manage multiple subagents Route tasks to appropriate agents Monitor progress and handle failures Aggregate results from multiple sources
Define workflow patterns and templates Implement error handling and recovery Manage state and progress tracking Coordinate handoffs between agents
Analyze task dependencies Create execution order Handle conditional execution Manage resource conflicts
Task A β Task B β Task C
Task A, Task B, Task C β Aggregate
Input β Task A β Task B β Task C β Output
Coordinator β Multiple Subagents β Results
Event β Trigger β Response β Next Event
orchestrate [workflow] - Execute complex workflow parallel [tasks] - Run tasks in parallel pipeline [steps] - Chain tasks in sequence supervise [agents] - Manage multiple agents dependencies [tasks] - Analyze and resolve dependencies
"Orchestrate a complete research project with multiple agents" "Run these tasks in parallel and combine results" "Create a pipeline for content creation from research to publication" "Supervise a team of agents working on different aspects" "Analyze dependencies and create execution order"
1. Research Topic β Research Agent 2. Data Collection β Data Agent 3. Analysis β Analysis Agent 4. Report Generation β Writing Agent 5. Review β QA Agent
1. Topic Research β Research Agent 2. Outline Creation β Writing Agent 3. Draft Writing β Writing Agent 4. Editing β Editing Agent 5. Publication β Publishing Agent
1. Requirements β Analysis Agent 2. Design β Design Agent 3. Implementation β Coding Agent 4. Testing β QA Agent 5. Deployment β Deployment Agent
sessions_spawn({ task: "specific task", label: "agent-name", mode: "run" })
subagents list
subagents kill [agent-id] subagents steer [agent-id] "new instructions"
Data Dependencies: Task B needs output from Task A Resource Dependencies: Tasks sharing same resources Order Dependencies: Tasks must run in specific order Conditional Dependencies: Task runs only if condition met
1. Identify all dependencies 2. Create dependency graph 3. Find topological sort 4. Execute in dependency order 5. Handle conflicts and cycles
Agent Failure: Subagent crashes or times out Dependency Failure: Required task fails Resource Conflict: Multiple agents need same resource Network Issues: API calls fail or timeout
Retry: Attempt failed task again Alternative: Use different approach or agent Skip: Continue without failed task Rollback: Undo previous steps
Track completed tasks Monitor current execution Record task results Maintain workflow state
Save progress at key points Enable restart from checkpoints Maintain consistency across failures
/sessions_send [agent-id] "instructions"
Auto-announce results Reply with findings Report errors and status
Share data through files Coordinate via shared state Trigger other agents
Identify independent tasks Run in parallel when possible Aggregate results efficiently
Monitor agent resource usage Balance load across agents Avoid resource conflicts
Task completion time Resource utilization Error rates Success rates
Max 10 concurrent subagents Max 2 levels of nesting 10-minute timeout per agent Automatic cleanup
Validate task inputs/outputs Maintain consistency Handle partial failures Ensure atomic operations
Main Coordinator β Team Coordinators β Individual Agents
Assign tasks based on agent capabilities Reassign if agent fails Balance load dynamically
Event β Trigger β Agent β Result β Next Event
Plan β Execute β Monitor β Adjust β Repeat
Use for complex self-improvement tasks Coordinate multiple evolution agents Manage long-term capability building
Orchestrate research projects Coordinate data analysis Manage multi-step investigations
Coordinate content production pipelines Manage multi-agent content creation Orchestrate publication workflows
# List running agents subagents list # Kill failed agent subagents kill [id] # Send instructions sessions_send [agent-id] "message" # Spawn new agent sessions_spawn({ task: "task", label: "name", mode: "run" })
# Research project orchestrate "research-project" with agents: research, analysis, writing # Content pipeline pipeline "content-creation" with steps: research, outline, draft, edit, publish # Software development supervise "dev-team" with agents: analysis, design, coding, testing, deployment
Start Simple: Begin with sequential execution Add Parallelism: Identify independent tasks Handle Failures: Implement robust error handling Monitor Progress: Track execution and results Optimize Performance: Balance load and resources
Task completion rate Execution time efficiency Resource utilization Error recovery effectiveness Overall workflow success Remember: Good orchestration makes complex tasks manageable and reliable.
Agent frameworks, memory systems, reasoning layers, and model-native orchestration.
Largest current source with strong distribution and engagement signals.