trimtab

Bootstrap repositories with self-verifying agentic workflows using player-coach verification.

1|1|Updated Mar 8, 2026
One-click install
npx skills add https://github.com/Hmbown/Trimtab --skill trimtab
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: trimtab
Source: https://github.com/Hmbown/Trimtab/tree/main/skills/trimtab
Command: npx skills add https://github.com/Hmbown/Trimtab --skill trimtab

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the setup and execution of a closed-loop, self-verifying agentic workflow within a repository, ensuring work is completed and independently verified.

Core Features & Use Cases

  • Workflow Bootstrapping: Scaffolds necessary files and configurations for a Trimtab-style agent workflow.
  • Agent Orchestration: Manages player/coach dynamics, typically with Claude as orchestrator and Codex as verifier.
  • Use Case: You want to establish a disciplined development process in a new project where AI agents perform tasks, and another AI agent independently verifies their work before anything is merged.

Quick Start

Use the trimtab skill to bootstrap a new repository with the self-verifying agent workflow.

Frequently Asked Questions about trimtab

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

FAQPage Schema
How do I set up a self-verifying agentic workflow in my repository?

To set up a self-verifying agentic workflow, you can use this Skill to bootstrap the necessary files and configurations. It scaffolds a Trimtab-style environment that enforces a player-coach model, managing task execution and independent verification before completion.

What is a player-coach model for AI development and how does it verify code?

A player-coach model for AI development uses one AI agent as an orchestrator to perform tasks and another as an independent verifier. This Skill automates that dynamic, typically using Claude to manage execution and Codex to package evidence and provide verification verdicts.

Can I use Claude and Codex together for automated task execution and verification?

Yes, you can use Claude and Codex together for automated task execution and verification. This Skill orchestrates them through a player-coach dynamic where Claude manages the workflow and Codex independently verifies the work, ensuring progress is checked before completion.

How do I bootstrap a CI/CD pipeline with independent AI verification?

To bootstrap a CI/CD pipeline with independent AI verification, this Skill scaffolds the required configurations in your repository. It establishes a closed-loop agentic workflow that manages task execution, packages evidence, and delivers verification verdicts across various tools.

Does this self-verifying agent workflow integrate with task management tools like Linear?

Yes, the self-verifying agent workflow integrates with task management tools like Linear. The Skill manages task execution and verification verdicts across various platforms, including Linear, Claude, and Codex, ensuring work is independently verified before merging.