AGENTS.md is a project-level configuration file that provides AI coding assistants with complete project context, development principles, and operational guidelines. It follows the agents.md standard and works across Claude, Cursor, GitHub Copilot, and other AI tools.

What is Jimmy's Workflow?

Jimmy's Workflow is a four-phase validation system (PRE-FLIGHT, IMPLEMENT, VALIDATE, CHECKPOINT) designed to prevent AI hallucination and ensure robust implementation. Each phase has specific gates and confidence levels (HIGH/MEDIUM/LOW) that determine when human review is needed.

How do I set up multiple AI instances as a team?

Assign each AI instance a role card with identity, responsibilities, personality traits, and success criteria. Use file-based handoff protocols for coordination. Personality orchestration at the system prompt level produces measurably better results than treating AI instances as generic tools.

CLAUDE.md

CLAUDE.md is the quick-reference companion to AGENTS.md. While AGENTS.md provides comprehensive project context, CLAUDE.md gives Claude Code (and other AI assistants) a compact summary of principles, commands, and critical rules.

What it contains

Section	Purpose
Core Principles	Numbered list of all 11 principles
Jimmy’s Workflow	Quick summary with phase diagram
Documentation Navigation	How to find and follow project docs
Template Compliance	How to run audit checks
Critical Rules	Do/don’t list for quick scanning
GitHub Operations	Common `gh` CLI commands
Common Commands	Project-specific dev/test/build commands

Why both AGENTS.md and CLAUDE.md?

AGENTS.md is the universal standard — any AI tool reads it. It’s comprehensive (600+ lines) with full project context, architecture, and troubleshooting.

CLAUDE.md is Claude Code-specific. It’s a compact (~125 lines) reference that:

Points to AGENTS.md as the authoritative source
Lists commands the AI can run immediately
Provides the critical rules at-a-glance
Includes template compliance audit triggers

Claude Code reads CLAUDE.md automatically on session start. Other AI tools read AGENTS.md. Both files work together.

The compliance audit trigger

CLAUDE.md includes a trigger pattern:

When user says “check templates” OR “check if we’re following the rules”:

Read AGENTS.md first

Follow the compliance audit workflow

Execute audit-project.sh

Generate a compliance report

Offer remediation

This means saying “check templates” in a Claude Code session automatically triggers a full project audit.

Filling in the template

CLAUDE.md has fewer placeholders than AGENTS.md. The main ones:

# Common Commands
[DEV_COMMAND]          # e.g., bun run dev
[TEST_COMMAND]         # e.g., bun test
[BUILD_COMMAND]        # e.g., bun run build
[TYPECHECK_COMMAND]    # e.g., bun run typecheck
[LINT_COMMAND]         # e.g., bun run lint

Replace these with your project’s actual commands.

Getting started

git clone https://github.com/ydun-code-library/Ydun_ai_workflow.git
cd Ydun_ai_workflow
cp templates/core/CLAUDE.md.template /path/to/your-project/CLAUDE.md
# Fill in [DEV_COMMAND], [TEST_COMMAND], etc.
# Verify: grep "\[.*COMMAND\]" CLAUDE.md