Execute independent multi-agent tasks in parallel with model-tier routing.
npx skills add https://github.com/xdkp/oh-my-claudecode --skill ultrawork-xdkp
Please help me install this Agent Skill. Skill: ultrawork Source: https://github.com/xdkp/oh-my-claudecode/tree/main/skills/ultrawork Command: npx skills add https://github.com/xdkp/oh-my-claudecode --skill ultrawork-xdkp
Ultrawork solves slow, sequential execution when you have multiple independent tasks that could be handled at the same time, wasting time and increasing overall runtime.
Ask your system to run ultrawork to delegate several independent subtasks concurrently, such as implementing an endpoint, adding integration tests, and updating related documentation in parallel.
High-intent search queries and answers about installing and using this skill.
Parallel Claude Code tasks run by firing independent calls simultaneously and explicitly passing the model for each delegation, reducing total completion time for concurrent work items while dependent steps remain sequenced.
Model-tier routing in Claude Code assigns each subtask to an appropriate model tier, routing simple lookups to Haiku, standard work to Sonnet, and complex analysis to Opus to balance execution speed and operational cost.
Multi-agent orchestration for independent tasks launches multiple agents simultaneously to handle work items with no dependencies between them, enabling parallel execution instead of slow sequential processing.
Parallel execution in Claude Code optionally uses background execution for long-running operations, applying lightweight completion checks to monitor progress without blocking other concurrent independent subtasks.
You should not use parallel execution for task delegation when steps have dependencies between them, because this approach requires independent work items to run concurrently while dependent steps must remain sequenced.
Parallel execution in Claude Code does not require orchestration modes to function, acting as a composable layer that is typically orchestrated by higher modes like ralph and autopilot without adding persistence loops.
prthikExecute independent subtasks in parallel across multiple agents.
sartoris-digitalExecute implementation plans with sequential, parallel, autopilot, and Ralph modes.
chanhee-kangExecute multiple agents in parallel with explicit model selection.
eszxcvfdOrchestrate parallel task execution across multiple agents with tier-aware routing.
