What is agent-thread AGENTS.md?

Repository instructions for a Next.js web app built and deployed with Bun and Cloudflare Workers.

Installing dependencies, running tests and builds, developing the app, and following its deployment workflow.

They prevent the project from drifting to incompatible tools such as npm, Vite, Jest, or Express.

agent-thread AGENTS.md — Instructions file by amalshaji

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/amalshaji/agent-thread/agents-md

Clone the repo

git clone --depth 1 https://github.com/amalshaji/agent-thread

Made for: Codex, OpenCode.

Per session 694 This file is loaded in full into every session.

When invoked 694 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.

Model	Per session	Once invoked
Fable 5	$0.00694	$0.00694
Opus 5	$0.00347	$0.00347
Sonnet 5	$0.00139	$0.00139
Haiku 4.5	$0.00069	$0.00069

Measured 2d ago against content hash 226ad11ed810, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-thread 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 2d 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 · 56 lines

How it starts

The opening of the file, as written. The whole thing — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Instructions

This repository is a Bun-managed Next.js/OpenNext app. Use Bun for project commands and dependency management, and keep changes aligned with the existing Next.js App Router and Cloudflare Worker deployment path.

Tooling

Use bun install instead of npm install, yarn install, or pnpm install.
Use bun run <script> instead of npm run <script>, yarn run <script>, or pnpm run <script>.
Use bun test for tests.
Use bunx <package> <command> instead of npx <package> <command>.
Do not introduce Vite, Express, Jest, Vitest, npm, yarn, or pnpm.
When a dependency exposes a Node-oriented CLI, invoke it through Bun, matching the existing scripts. For example: bun --bun ./node_modules/next/dist/bin/next build.
Codex or editor host tooling may use its own configured Node runtime when that tool requires it, such as js_repl or browser automation. That exception does not justify adding Node setup, npm scripts, or a project-level Node workflow.

Project Commands

Install dependencies: bun install
Run the Next.js dev server: bun run dev
Build the app and CLI: bun run build
Build only the CLI: bun run build:cli
Preview the Cloudflare Worker: bun run preview
Deploy through OpenNext/Cloudflare: bun run deploy
Type-check: bun run check
Run tests: bun test
Run the local CLI entrypoint: bun run cli

Release Workflow

When updating the CLI version, title the pull request Release cli {version}, replacing {version} with the exact version being released.

Architecture

The frontend and API routes live in Next.js App Router.
The Cloudflare Worker is produced by @opennextjs/cloudflare; wrangler.toml points at .open-next/worker.js.
Keep next.config.ts compatible with the OpenNext build path, including output: "standalone", unless the deployment strategy is intentionally changed.
Keep server-side rendering and public transcript rendering in the Next app rather than reintroducing a separate Bun.serve frontend.
Keep Cloudflare bindings and runtime-specific code isolated behind the existing helper modules.

Read the full file on GitHub · 56 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.

2d ago First seen · 56 lines · 694 tokens per session scan A 226ad11ed810

Subscribe to this mod's changes

agent-thread AGENTS.md is an instructions file published in the GitHub repository amalshaji/agent-thread (1 stars, last pushed 3mo ago), licensed MIT. It adds 694 tokens to every session, about $0.0035 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 2026-08-31.

Frontend Git and workflow Testing agent-tools bun claude-code cloudflare-workers codex-cli nextjs

agent-thread AGENTS.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Agent Instructions

Tooling

Project Commands

Release Workflow

Architecture

What this file has done since we first saw it

Other instructions, from other repositories

codex_remote AGENTS.md

cloudflare-workers-nextjs-saas-template AGENTS.md

open-connector AGENTS.md

munkel CLAUDE.md

codex-studio AGENTS.md

titen AGENTS.md