Bolt vs OpenAI Codex CLIUpdated 2026-03-16

SUMMARY

Bolt (Platform) uses text configuration while OpenAI Codex CLI (CLI) uses markdown. They differ on 6 of 11 compared features. Both tools provide AI-assisted development but take different approaches to project configuration.

FEATURE COMPARISON

FeatureBoltOpenAI Codex CLI
Config formattextmarkdown
Hierarchy support
Global config
Project config
Subdirectory scoping
File inclusion / imports
Ignore file.codexignore
IDE integrationPlatformCLI
Schema / structureFree-form (Markdown)Free-form (Markdown)
Git committed
EncodingUTF-8UTF-8

FILE MAPPING

PurposeBoltOpenAI Codex CLI
Project instructions.bolt/promptAGENTS.md
Project instructionsAGENTS.mdAGENTS.override.md

SIDE-BY-SIDE CODE SAMPLES

Bolt
.bolt/prompt
text
You are building a SaaS dashboard application.

Stack: Remix with TypeScript, Prisma ORM, PostgreSQL, TailwindCSS.

Key rules:
- Use server loaders for data fetching, not client-side fetch
- All forms use Remix Form with progressive enhancement
- Use Prisma for all database operations
- Style with TailwindCSS utility classes, no custom CSS
- Components go in app/components/, routes in app/routes/
- Use zod for form validation on both client and server

OpenAI Codex CLI
AGENTS.md
markdown
# AGENTS.md

## Project Overview

This is a Next.js 14 e-commerce application using the App Router,
Server Components, and Drizzle ORM with PostgreSQL.

## Code Conventions

- Use TypeScript strict mode throughout
- Prefer Server Components; use "use client" only when necessary
- Name files with kebab-case: `product-card.tsx`, not `ProductCard.tsx`
- Co-locate tests: `product-card.test.tsx` next to `product-card.tsx`

## Architecture

- `app/`           — Next.js App Router pages and layouts
- `components/`    — Shared UI components
- `lib/`           — Utilities, database client, auth helpers
- `actions/`       — Server Actions for mutations

## Testing

- Use Vitest for unit tests, Playwright for E2E
- Mock database calls using `lib/test-utils.ts` helpers
- Every new component needs at least one test

## Build Commands

```bash
pnpm dev          # Start dev server
pnpm build        # Production build
pnpm test         # Run Vitest
pnpm test:e2e     # Run Playwright
pnpm lint         # ESLint check
```

KEY DIFFERENCES

1. Config format

Bolt: text. OpenAI Codex CLI: markdown.

2. Hierarchy support

Bolt: Not supported. OpenAI Codex CLI: Supported.

3. Global config

Bolt: Not supported. OpenAI Codex CLI: Supported.

4. Subdirectory scoping

Bolt: Not supported. OpenAI Codex CLI: Supported.

5. Ignore file

Bolt: Not supported. OpenAI Codex CLI: .codexignore.

6. IDE integration

Bolt: Platform. OpenAI Codex CLI: CLI.

WHICH SHOULD I USE?

Recommendation

Choose based on your IDE/CLI preference. If you use Bolt, configure .bolt/prompt. If you use OpenAI Codex CLI, configure AGENTS.md. Many projects include config files for multiple tools so each team member can use their preferred editor.

PORTABILITY TIP

sync-instructions.sh
bash
# Maintain a single source of truth:
cp AI-INSTRUCTIONS.md CLAUDE.md
cp AI-INSTRUCTIONS.md .cursorrules
cp AI-INSTRUCTIONS.md AGENTS.md

SEE ALSO

agentconfig.ing2026-03-16COMPARE(1)