create-better-t-stack AGENTS.md

Repository instructions for a Bun and Turborepo monorepo, meaning one repository that contains multiple related applications and packages. They describe its folders, commands, coding style, and error-handling practices.

In plain words
What is it for?
Developing and checking the create-better-t-stack project, including its command-line app, Next.js website, shared packages, backend, and installable agent plugin.
Why use it?
They give contributors a shared way to build, test, format, and organize the CLI, web app, backend, templates, types, and plugin.

Instructions file for CodexOpenCode

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/amanvarshney01/create-better-t-stack/agents-md
Clone the repo
git clone --depth 1 https://github.com/AmanVarshney01/create-better-t-stack

Made for: Codex, OpenCode.

Per session 938 This file is loaded in full into every session.
When invoked 938 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00938 $0.00938
Opus 5 $0.00469 $0.00469
Sonnet 5 $0.00188 $0.00188
Haiku 4.5 $0.00094 $0.00094

Measured 3d ago against content hash e828b642dab6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

create-better-t-stack AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

AGENTS.md · 81 lines

How it starts

The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Repository Guidelines

Project Structure & Module Organization

This repo is a Bun + Turborepo monorepo.

  • apps/cli: published CLI (create-better-t-stack), with source in apps/cli/src and tests in apps/cli/test.
  • apps/web: Next.js docs/site (apps/web/src, apps/web/content/docs, apps/web/public).
  • packages/template-generator: template generation engine used by the CLI.
  • packages/types: shared schemas/types.
  • packages/backend: Convex backend used by web features.
  • plugin/: installable agent plugin (Claude Code + Codex) — skills, commands, and MCP wiring; published via the root marketplaces (.claude-plugin/, .agents/plugins/).

Build, Test, and Development Commands

  • bun install: install workspace dependencies.
  • bun dev:cli: watch-build CLI package.
  • bun dev:web: run web app locally (next dev --port 3333).
  • bun build: build all packages/apps through Turbo.
  • bun build:cli: build only the CLI target.
  • bun run check: format + lint (oxfmt . && oxlint .).
  • cd apps/cli && bun run test: run CLI tests.

Coding Style & Naming Conventions

  • Language: TypeScript (strict mode enabled across projects).
  • Modules: ESM-first ("type": "module" where applicable).
  • Formatting/linting: oxfmt and oxlint; run bun run check before committing.
  • File naming: prefer kebab-case files (for example database-setup.ts).
  • Symbols: camelCase for functions/variables, PascalCase for types/components.
  • Keep feature logic near domain folders (helpers, utils, template-handlers).

Error Handling Conventions

  • In CLI code, prefer better-result over ad-hoc try/catch for recoverable flows.
  • Return typed Result<T, E> and use Result.ok, Result.err, Result.try, and Result.tryPromise.
  • Reuse domain errors from apps/cli/src/utils/errors.ts (CLIError, ProjectCreationError, UserCancelledError) and convert thrown prompt errors at boundaries.

Template Authoring (Handlebars)

  • Templates live in packages/template-generator/templates and use helpers from packages/template-generator/src/core/template-processor.ts (eq, ne, and, or, includes).
  • For conditional ORM-specific output, use helper form with quoted values:
    • {{#if (eq orm "prisma")}}
    • {{else if (eq orm "drizzle")}}
    • {{/if}}
    • Example: packages/template-generator/templates/packages/infra/alchemy.run.ts.hbs.
  • When files must contain literal {{ ... }} (Vue/JSX/template syntax), escape opening braces as \{{ in .hbs files so Handlebars does not evaluate them.
    • Example: packages/template-generator/templates/frontend/nuxt/app/pages/index.vue.hbs.

Read the full file on GitHub · 81 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 81 lines · 938 tokens per session scan A e828b642dab6

Subscribe to this mod's changes

create-better-t-stack AGENTS.md is an instructions file published in the GitHub repository AmanVarshney01/create-better-t-stack (5,691 stars, last pushed 2d ago), licensed MIT. It adds 938 tokens to every session, about $0.0047 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.