Verified:GitHub

teamLimit

Configure Manager and Sub AI concurrency limits in workspace files.

Updated May 21, 2026

Search Tags:#state management #workflow orchestration #concurrency #configuration #ai capacity #resource throttling #parallel task execution

Ready to run

npx skills add https://github.com/kinetas/harness_engineering --skill teamlimit

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: teamLimit
Source: https://github.com/kinetas/harness_engineering/tree/main/teamLimit
Command: npx skills add https://github.com/kinetas/harness_engineering --skill teamlimit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents overloaded runs by letting you control how many Manager and Sub AI instances can execute concurrently in your workspace.

Core Features & Use Cases

Interactive capacity configuration: Prompts you to set Manager AI maximum concurrency and the Sub AI limit per Manager.
State-aware validation: Reads existing state and warns you when the new Manager limit is below currently active managers to avoid disruptive behavior.
Persistent updates to operational files: Updates both company state and AI list capacity sections so future task scheduling respects the new limits.
Use case: You run a large project and want to cap concurrency to reduce cost or stabilize performance while tasks are partitioned into segments.

Quick Start

Invoke /teamLimit and enter the desired Manager limit and Sub AI per Manager values when prompted.

Frequently Asked Questions about teamLimit

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

FAQPage Schema

How do I limit concurrent AI execution in my workspace?▼

You can set parallel task execution limits by configuring Manager and Sub AI capacity values. This caps concurrent AI execution to prevent overloaded workspace runs and stabilize performance during large project scheduling.

How do I configure state management limits for parallel task execution?▼

You configure state management limits by invoking the configuration tool and entering desired Manager limit and Sub AI per Manager values. The system reads existing state files, validates minimum values, and updates both files with new concurrency caps.

What happens if I reduce the Manager limit below currently active managers?▼

If you reduce the Manager limit below currently active managers, the system issues a state-aware validation warning. This alerts you to potentially disruptive behavior before applying the lower concurrency cap to your running workspace.

Do I need specific state files to apply resource throttling to AI workflows?▼

Yes, resource throttling requires reading doc/company_state.json and doc/AI_list.txt. The tool validates existing state from these files and persistently updates both operational files with the new managerLimit and subAILimit values.

When should I use concurrency limits for workflow orchestration?▼

Use concurrency limits for workflow orchestration when running large projects requiring cost reduction or performance stabilization. Capping parallelism manages resource throttling while tasks are partitioned into segments for scheduling.