What is Always-On Guidance?

It is a specialized skill for Claude Code that enforces behavioral rules, tool usage standards, and model-specific defaults for agents working in a swarm environment.

Does this skill help with automated testing?

Yes, it mandates a strict Red-Green-Refactor testing discipline whenever tests cover the area being modified by the agent swarm.

How does it handle different AI models?

It provides tailored instructions for GPT-5.2-code, focusing on brevity and imperatives, and Opus 4.5, allowing for brief rationales and phased planning.

How does it prevent code conflicts in a swarm?

The skill enforces a file reservation system using time-to-live (TTL) locks to ensure only one agent edits a specific file at a time, releasing the lock upon task completion.

Always-On Swarm Guidance

Name: Always-On Swarm Guidance
Author: primeinc

byprimeinc

0•

Productivity & Workflow

Enforces consistent behavior, tool protocols, and model-specific execution patterns across autonomous agent swarms.

Always-On Swarm Guidance provides a standardized operational framework for Claude-plugin agents operating within a multi-agent swarm architecture. It ensures seamless coordination by managing file reservations, subtask handoffs, and tool selection while optimizing output styles for specific model architectures like GPT-5.2-code and Opus 4.5. By implementing strict instruction priorities and testing disciplines, this skill minimizes errors in complex environments and ensures all agents adhere to the latest hive and swarm protocols while avoiding deprecated legacy references.

Key Features

01Strict Instruction Priority Management

02Universal Swarm Tool Integration (hive, swarm, hivemind)

03Standardized Red-Green-Refactor Testing Discipline

04Model-Specific Behavior Optimization for GPT and Opus

050 GitHub stars

06Automated File Reservation and Conflict Prevention

Use Cases

01Managing complex multi-agent workflows to prevent file conflicts during concurrent edits.

02Optimizing agent response styles based on the specific strengths of the underlying LLM.

03Standardizing testing and coordination protocols across automated software engineering tasks.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add primeinc/swarm always-on-guidance

For use in Claude.ai and ChatGPT

Download Skill