Orchestrate multiple Claude Code sessions for parallel agent tasks.
npx skills add https://github.com/nsheaps/ai-mktpl --skill agent-teams-nsheaps
Please help me install this Agent Skill. Skill: agent-teams Source: https://github.com/nsheaps/ai-mktpl/tree/main/.ai/plugins/agent-teams-skills/skills/agent-teams Command: npx skills add https://github.com/nsheaps/ai-mktpl --skill agent-teams-nsheaps
Agent teams simplify the complexity of running many coordinated Claude Code instances in parallel by providing spawn, communication, and lifecycle controls so leads can manage multiple independent agents without manual orchestration.
Start an agent team to run three parallel reviewers on different modules and require plan approval before merging.
High-intent search queries and answers about installing and using this skill.
You orchestrate parallel Claude Code sessions by spawning multiple agents as teammates, where each independently inspects a different module and reports findings to a lead for consolidation. This multi-agent workflow handles concurrent task execution without manual coordination.
Multi-agent orchestration coordinates independent agents executing separate tasks concurrently. It manages task lifecycles through shared mailboxes, direct peer messaging, dependency tracking, and automatic unblocking to ensure seamless parallel execution across all spawned agents.
Yes, you can launch teammates in tmux-backed panes to display and manage parallel Claude Code agents. You can also run them in-process, allowing you to persist or hide sessions as needed for flexible multi-agent workflow control.
Hooks handle task completion and idle states by enabling script-based validations through TeammateIdle and TaskCompleted events. These safety gates ensure specific conditions are met before agents transition to idle or complete their assigned tasks.
Environment variable propagation is automatically handled for spawned agents in Claude Code. This ensures all parallel teammates inherit the necessary configuration and context from the lead to execute their independent tasks concurrently and consistently.
The best way to coordinate independent agents for modular testing is using multi-agent orchestration with task dependency tracking and shared mailboxes. This ensures parallel test execution across separate modules with automatic unblocking and lead-managed consolidation.
dakshrawat298-gifControl tmux sessions by sending keystrokes and capturing pane output.
mor-duongmhDispatch parallel agents with isolated context to investigate failures and consolidate remediation plans.
ramonclaudioCoordinate Claude Code sessions into a collaborative agent team with task graphs.
arthtyagiProvision Vercel Sandboxes to run benchmark scenarios and generate coverage reports.
