omc-reference

Catalog OMC agents, tools, pipeline stages, and git trailer conventions.

1|Updated Apr 1, 2026
One-click install
npx skills add https://github.com/OliverOuyang/shuhe-work-skills --skill omc-reference-oliverouyang
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: omc-reference
Source: https://github.com/OliverOuyang/shuhe-work-skills/tree/main/skills/omc-reference
Command: npx skills add https://github.com/OliverOuyang/shuhe-work-skills --skill omc-reference-oliverouyang

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides a single, discoverable reference for OMC agents, available tools, team pipeline stages, commit message conventions, and built-in skill invocation so teams do not need to duplicate orchestration guidance across CLAUDE.md sessions.

Core Features & Use Cases

  • Agent Catalog: Lists agent names, short personas, and recommended model routing for different tasks (explore, analyst, planner, executor, verifier, etc.).
  • Tools & Runtime APIs: Documents runtime capabilities (team APIs, notepad, project memory, state ops, LSP and AST helpers) and examples of invocation patterns.
  • Team Pipeline & Commit Protocol: Defines pipeline stages, remediation loops, and git trailer conventions to preserve decision context in commits.
  • Use Case: On delegating work to an agent or orchestrating a team, consult this reference to pick the right agent, tool, and commit format for reproducible, auditable automation.

Quick Start

Load the omc-reference to display the agent catalog, model routing tips, available runtime tools, and commit trailer guidance when planning or delegating OMC workflows.

Frequently Asked Questions about omc-reference

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

FAQPage Schema
What is agent orchestration and how does a centralized reference help manage AI workflows?

Agent orchestration coordinates delegating tasks to specialized AI agents across team pipelines. A centralized reference eliminates duplicated guidance across sessions by providing a single catalog for agent identities, tool APIs, and commit conventions to ensure reproducible automation.

How do I pick the right AI agent and model routing for different development tasks?

To pick the right AI agent and model routing, consult the agent catalog which lists short personas and recommended model routing for specific tasks like exploring, analyzing, planning, executing, and verifying code in collaborative development workflows.

What git trailer format should I use for commit messages in automated team pipelines?

Git trailer formats for automated team pipelines preserve decision context in commits during collaborative development. The centralized reference defines the specific commit conventions and trailer structures required to maintain reproducible and auditable orchestration.

Can I use runtime tools like notepad and state ops during agent delegation workflows?

Yes, you can use runtime tools including notepad, project memory, state ops, LSP, and AST helpers during agent delegation. The reference documents these runtime capabilities and provides invocation pattern examples for orchestrating team workflows.

Does this orchestration reference require external dependencies or specific runtime environments?

No external dependencies are required to load the orchestration reference. It operates as a standalone centralized catalog that can be applied directly to AI orchestration scenarios for agent delegation, team routing, and workflow invocation without prerequisite packages.

Best way to define team pipeline stages and remediation loops for AI agents?

The best way to define team pipeline stages and remediation loops is using a centralized commit protocol. This reference outlines pipeline stages and remediation processes to preserve decision context and ensure auditable automation across collaborative development.