takt

Orchestrate multi-agent workflows from piece YAML with a Team Lead.

1|Updated Jan 5, 2026
One-click install
npx skills add https://github.com/j5ik2o/dotfiles --skill takt-j5ik2o
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: takt
Source: https://github.com/j5ik2o/dotfiles/tree/main/references/takt/builtins/skill
Command: npx skills add https://github.com/j5ik2o/dotfiles --skill takt-j5ik2o

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

TAKT automates multi-agent orchestration using a Team Lead to coordinate agent movements according to a piece YAML workflow.

Core Features & Use Cases

  • Sequential and parallel movement orchestration defined by piece YAML.
  • Resolution of persona, policy, instructions, knowledge, and output contracts to build executable prompts for agent tasks.
  • Loop monitoring, report handling, and robust completion/abort transitions for resilient run management.

Quick Start

Configure TAKT with your piece YAML and run the first movement to begin orchestration.

Frequently Asked Questions about takt

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

FAQPage Schema
How do I orchestrate multi-agent workflows with parallel and sequential execution?

Multi-agent workflow orchestration is automated using a Team Lead to coordinate agent movements defined by a piece YAML workflow, supporting both sequential and parallel tasks with loop monitoring.

What is a piece YAML workflow in multi-agent orchestration?

A piece YAML workflow defines sequential and parallel movements for a Team Lead to coordinate, resolving persona, policy, instructions, knowledge, and output contracts to generate executable prompts for agent tasks.

How do I generate executable prompts for agent tasks in a multi-agent system?

Executable prompts are generated by resolving persona, policy, instructions, knowledge, and output contracts defined in the workflow, ensuring each agent receives structured input for task execution.

Can I use a Team Lead to coordinate parallel agent movements and handle reports?

A Team Lead coordinates parallel agent movements, initializes runs, spawns Task tool agents, evaluates rules, and handles reports to gracefully complete or abort runs on completion or error.

When should I use YAML-based multi-agent orchestration instead of single agent execution?

YAML-based multi-agent orchestration is ideal when tasks require coordinated sequential and parallel movements, loop monitoring, and strict resolution of policies and output contracts across multiple agents.

How does multi-agent orchestration handle run errors and abort transitions?

Multi-agent orchestration manages run errors by evaluating rules, handling reports, and executing graceful completion or abort transitions to ensure resilient run management upon task failure.