agent-team-codex-harness

Design, build, and audit Codex-native harnesses for Agent Team projects.

2|Updated May 10, 2026
One-click install
npx skills add https://github.com/tae2089/agent-team --skill agent-team-codex-harness
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-team-codex-harness
Source: https://github.com/tae2089/agent-team/tree/main/skills/agent-team-codex-harness
Command: npx skills add https://github.com/tae2089/agent-team --skill agent-team-codex-harness

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides a structured, Codex-native harness framework for building, evolving, and auditing Agent Team harnesses. It defines how to assemble reusable components such as .codex/agents, .agents/skills, an orchestrator, and AGENTS.md pointers, plus references, to enable durable, daemonless runtime coordination and continuous evolution.

Core Features & Use Cases

  • Scaffold and standardize Codex harness components (agents, skills, orchestrator) and ensure discoverability.
  • Provide governance for reference loading, runtime state coordination, and artifact-driven workflows.
  • Support harness auditing, evolution, and follow-up work through AGENTS.md pointers and references.
  • Use cases include building new domain harnesses, auditing existing deployments, and extending with QA and orchestration patterns.

Quick Start

Review the SKILL.md to set up an initial harness using the orchestrator and runtime skills, then run through a sample audit workflow.

Frequently Asked Questions about agent-team-codex-harness

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

FAQPage Schema
How do I orchestrate a daemonless runtime for agent teams?

You can orchestrate a daemonless runtime by using a Codex harness to assemble reusable components like an orchestrator, agent templates, and AGENTS.md pointers for durable coordination without a background daemon.

What is a Codex harness and how does it manage agent workflows?

A Codex harness is a structured framework that manages agent workflows by applying contract-driven logic to load frontmatter, references, and required helper skills with explicit data flow and error handling.

How do I scaffold and standardize components for agent team projects?

Scaffold and standardize components by assembling .codex/agents, .agents/skills, and an orchestrator, ensuring discoverability and governance through AGENTS.md pointers and reference loading.

Can I audit existing agent harness deployments and extend their orchestration patterns?

Yes, you can audit existing deployments and extend orchestration patterns by using the harness framework to evaluate artifact-driven workflows, support continuous evolution, and trigger follow-up work via references.

Does the agent team orchestrator support contract-driven workflows with explicit error handling?

Yes, the orchestrator supports contract-driven workflows by specifying the loading of frontmatter and references, ensuring explicit data flow, runtime state coordination, and structured error handling.

When do I need AGENTS.md pointers for runtime state coordination?

You need AGENTS.md pointers when setting up durable, daemonless runtime coordination to ensure artifact-driven workflows, reference loading governance, and continuous harness evolution across your agent team project.