AI Skill StudioAI Skill Studio
Skills
Presets:

Target Standard & File Format

AI Agent Rules & Skills
Context Shields & Privacy (Ignore Files)
Integrations & Machine Documentation
AI Governance & Multi-Agent Planning Suite

Identity & Activation Rules

3 chips
Scoped domain keywords evaluated in real time
api-routescode-auditsrefactoring
Suggested Chips:

Technology Stack Context

Engineering Philosophy

Flexible & Adaptable

Architectural & Code Quality Conventions

Less rigid & configurable

Agent Behavioral Guardrails

Select active rules
Security & Trust Boundaries
Always included
  • Untrusted Content Is Data
  • Encoded Payloads
  • Least-Privilege File Access
  • Confirm Irreversible Actions
  • Secret Handling
  • Validation Failures
  • Conflicting Requests

Step-by-Step Workflow Procedures

StarARCHITECTURE.MD
Specialist Persona
ARCHITECTURE.md Specification Hub Directory

Commit Messages & Pull Requests ARCHITECTURE.md Specification

Generate an AI-optimized ARCHITECTURE.md invariant specification for Commit Messages & Pull Requests. Documenting module responsibilities, state boundaries, and data flow in ARCHITECTURE.md provides autonomous coding assistants with high-level structural invariants before any code modifications begin.

Target Path
ARCHITECTURE.md
Execution Scope
Read when referenced
Specification Format
Markdown (.md)
AI Tool Support
Any coding agent
02 / WHY IT HELPS

What this file changes for Commit Messages & Pull Requests

Without it

Documenting module responsibilities, state boundaries, and data flow in ARCHITECTURE.md provides autonomous coding assistants with high-level structural invariants before any code modifications begin.

What the file tells the agent
Read git status, the staged diff, and recent log style before writing.
Subject as type(scope): summary, imperative, at most 72 characters.
Propose separate commits when changes mix unrelated concerns.
Describe only what the diff contains; never claim unverified tests.
No commit, amend, or push without approval; never --no-verify or force-push.
04 / REPOSITORY PLACEMENT & 3-STEP TERMINAL INSTALLATION

How to Install Commit Messages & Pull Requests ARCHITECTURE.md Specification via Terminal

1

Step 1: Open Project Directory & Verify Target Placement

Open your terminal and navigate to your project root folder where the ARCHITECTURE.md file will reside. Ensure the file is placed at the exact path below relative to your project root so the AI engine automatically loads it:

ARCHITECTURE.md
2

Step 2: Fetch Rule File via Terminal Command

Run curl, PowerShell, or wget to stream the rule directly from the DevScratchpad raw API endpoint and write it to ARCHITECTURE.md:

Terminal One-Liner Install

Run directly in your project root to stream and write this rule file with one command.

Raw API Stream
$curl -fsSL "https://www.devscratchpad.tech/api/raw/architecture-md/commit-pr-writer" -o "ARCHITECTURE.md"
3

Step 3: Check that Any coding agent uses it

Agents read it when your CLAUDE.md or AGENTS.md points to it. Add a line such as "Read ARCHITECTURE.md before structural changes".

05 / ROUTE DIRECTORY & CROSS-TOOLING
Format Pillar Hub

ARCHITECTURE.md generator: a codemap and invariants for your repo

ARCHITECTURE.md Specification: how it works, where the file goes, and every ARCHITECTURE.md preset.

/architecture-md Directory
06 / RELATED DEVELOPER TOOLS

Tools for Commit Messages & Pull Requests work

Browser-based utilities to test, format and inspect the code your AI tool writes. Nothing you paste is uploaded.

View all 28 tools

Live Repository & Schema Ingestion

Inspect any public or private GitHub repository, or paste SQL DDL and Prisma models directly in AI Skill Studio to auto-generate authentic rules. Zero server transmission.

Launch AI Skill Studio

Migrating from Legacy .cursorrules or CLAUDE.md?

Use the free, in-browser converter to turn one long rules file into Cursor .mdc, Claude SKILL.md or Windsurf rules.

Convert Rules Free
07 / FREQUENTLY ASKED QUESTIONS

Technical FAQ: Commit Messages & Pull Requests AI Rulebooks

What belongs in ARCHITECTURE.md?

High-level component relationships, state flow, dependency graphs, and technical invariants that AI agents must never violate.

How does ARCHITECTURE.md help AI agents?

It tells them where code belongs and which boundaries to keep, so they are less likely to add circular dependencies or skip layers. Point to it from CLAUDE.md or AGENTS.md so they read it.