ck:team

Orchestrate parallel multi-session agent collaboration with shared task coordination and summary reports.

Updated Mar 7, 2026
One-click install
npx skills add https://github.com/MinhHoangDono/antigravity-kit --skill ck-team-minhhoangdono
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ck:team
Source: https://github.com/MinhHoangDono/antigravity-kit/tree/main/.agent/skills/team
Command: npx skills add https://github.com/MinhHoangDono/antigravity-kit --skill ck-team-minhhoangdono

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It coordinates multiple independent AI sessions in parallel so complex research, implementation, review, and debugging can happen concurrently without losing structure or accountability.

Core Features & Use Cases

  • Parallel multi-role collaboration: run independent researcher, developer, reviewer, and debugger teammates with isolated context windows.
  • Template-driven workflows: use built-in team modes for research, cook (plan → build → test → finalize), code review, and adversarial debugging.
  • Shared coordination protocol: create tasks, spawn teammates with required team_name, monitor completion, synthesize outputs into summary reports, then shut down cleanly.

Quick Start

Run ck:team cook with your plan description so the toolkit orchestrates developers and a tester to implement, verify, and finalize the work.

Frequently Asked Questions about ck:team

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I coordinate parallel agent teams for concurrent code review and debugging?

Parallel agent teams coordinate concurrent code review and debugging by spawning independent teammate sessions with isolated contexts, monitoring progress via events, and synthesizing outputs into summary reports. This preserves structure and accountability across independent workstreams.

What is the best way to orchestrate multi-session AI agents for research synthesis?

Multi-session AI agent orchestration for research synthesis uses template-driven workflows to spawn independent researcher teammates. It manages shared task coordination, monitors completion events, and generates summary reports before safely shutting down the team.

How do I run parallel implementation tasks with isolated agent context windows?

Parallel implementation tasks run by applying the cook template to orchestrate developers and a tester. Teammates spawn with isolated context windows to plan, build, test, and finalize the work concurrently, followed by clean shutdown and cleanup.

Do I need Agent Teams enablement to run parallel multi-role agent collaboration?

Yes, Agent Teams enablement is required to run parallel multi-role agent collaboration. The orchestration executes template-specific TeamCreate, TaskCreate, teammate spawn with team_name, progress monitoring, report generation, and TeamDelete commands.

Can I use task coordination to monitor concurrent workstreams and generate reports?

Task coordination monitors concurrent workstreams through event-driven monitoring of spawned teammates. It tracks completion across independent sessions and synthesizes their isolated outputs into a shared summary report before executing safe team deletion.

What are the limitations of using parallel orchestration for adversarial debugging?

Parallel orchestration for adversarial debugging is limited by the need for template-specific workflows and Agent Teams enablement. It requires explicit TeamCreate and TeamDelete execution, meaning workflows lacking safe shutdown protocols may lose structure or accountability.