team-agents

Coordinate parallel agent teams for complex tasks with dependencies and status tracking.

1|Updated May 24, 2026
One-click install
npx skills add https://github.com/tinner-deinno/innova-skills-lib --skill team-agents-tinner-deinno
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: team-agents
Source: https://github.com/tinner-deinno/innova-skills-lib/tree/main/core/team-agents
Command: npx skills add https://github.com/tinner-deinno/innova-skills-lib --skill team-agents-tinner-deinno

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

team-agents solves the challenge of breaking a single complex task into parallel work that still stays coordinated, so you can get multiple specialists producing results with a unified final output.

Core Features & Use Cases

  • Coordinated agent teams: Spin up coordinated teammates using TeamCreate/SendMessage/TaskList patterns instead of relying on independent subagents.
  • Structured task lifecycle: Create tasks with dependencies, collect progress via a heartbeat protocol, and compile team reports into a single synthesized document.
  • Worktree-based parallelism: Optionally isolate each agent with its own git worktree/branch to prevent file overwrites and enable clean merging.
  • Operational controls: Provide presence/status commands (who, sync, merge, compile), plus cleanup/doctor utilities for tmux-based teammate sessions.

Quick Start

Use team-agents by telling the AI: /team-agents "review this PR for security, performance, and tests" --roles 3 --worktree --plan.

Frequently Asked Questions about team-agents

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

FAQPage Schema
How do I coordinate multiple AI agents to review a pull request in parallel?

You coordinate multiple agents by spinning up teammates with TeamCreate and TaskList patterns, assigning roles for security, performance, and tests. Progress is collected via a heartbeat protocol and compiled into a single synthesized review document.

How does git worktree isolation work when running parallel coding agents?

Git worktree isolation optionally assigns each agent its own branch and working directory to prevent file overwrites during concurrent operations. This enables clean merging of individual agent outputs back into the main codebase without conflicts.

Can I use parallel agents for large code refactoring with task dependencies?

Yes, coordinated agent teams support complex refactors requiring multiple specialists. The structured task lifecycle creates tasks with dependencies, tracks status via heartbeats, and compiles team reports into a unified document for dependency-aware refactoring.

What is the best way to manage tmux sessions for background AI coding agents?

Operational controls provide presence and status commands like who, sync, merge, and compile, alongside cleanup and doctor utilities for tmux-based teammate sessions. These manage background agents and monitor their progress throughout the workflow.

Do I need independent subagents or coordinated teammates for complex research workflows?

Coordinated teammates are required when your research needs a unified outcome with dependency tracking and final report compilation. Unlike independent subagents, coordinated teams use SendMessage orchestration and heartbeat progress reporting to synthesize results.