cloudbox AGENTS.md

A project guide for Cloudbox, an open-source Cloudflare service that leads visitors from its homepage to a demo, documentation, source code, and deployment. It records the approved commands, deployment process, account boundaries, and production secrets.

In plain words
What is it for?
Running builds, type checks, tests, demos, and end-to-end checks; preparing production deployments through GitHub Actions; and verifying Cloudflare tokens, accounts, databases, and worker scripts.
Why use it?
It gives coding agents one reliable set of project rules and helps prevent deploying to the wrong Cloudflare account or missing required production configuration.

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/acoyfellow/cloudbox/agents-md
Clone the repo
git clone --depth 1 https://github.com/acoyfellow/cloudbox

Made for: Codex, OpenCode.

Per session 670 This file is loaded in full into every session.
When invoked 670 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 $0.00670 $0.00670
Opus 5 $0.00335 $0.00335
Sonnet 5 $0.00134 $0.00134
Haiku 4.5 $0.00067 $0.00067

Measured 2d ago against content hash 887b7a892b9a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cloudbox 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 · 125 lines

How it starts

The opening of the file, as written. The whole thing — 125 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Cloudbox agent guide

Cloudbox is an OSS Cloudflare funnel: homepage → demo → docs → source → Deploy to Cloudflare.

Canonical commands

bun run build
bun run typecheck
bun run test
bun run demo:fast
bun run demo:browser
CLOUDBOX_E2E_URL=https://cloudbox.coey.dev node scripts/e2e.mjs

Deploy model

Production deploys through GitHub Actions with Alchemy.

Canonical deploy file:

alchemy.run.ts

Production target:

https://cloudbox.coey.dev

Do not switch local Wrangler accounts to deploy Cloudbox. Use GitHub Actions secrets for production.

Cloudflare account boundaries

Cloudbox production belongs in the personal Cloudflare account:

[email protected]'s Account
bfcb6ac5b3ceaf42a09607f6f7925823

Do not deploy Cloudbox to the employee/AX account:

Agent Experience
31b91e7f9954ad8aa334d46f012bd8ed

Required production secrets

Set on the GitHub production environment:

CLOUDFLARE_ACCOUNT_ID
CLOUDFLARE_API_TOKEN
ALCHEMY_PASSWORD
ALCHEMY_STATE_TOKEN
CLOUDBOX_API_TOKEN
CLOUDBOX_D1_DATABASE_ID

Token must be validated against:

/user/tokens/verify
/accounts
/accounts/:account_id/d1/database?name=cloudbox-prod
/accounts/:account_id/workers/scripts/alchemy-state-store/script-settings
/accounts/:account_id/secrets_store/stores

Public copy rules

  • Do not mention private internal project names.
  • Do not use the word previously used for self-testing; call it demo, proof run, sample run, or launch-readiness run.
  • Public concept is: clean remote computer.
  • Public funnel is: demo → docs → GitHub → Deploy to Cloudflare.

Current vs next

Current:

  • deployed Astro/Worker app
  • demo workspace
  • receipts/artifacts/grading
  • local cloudbox.run() proof slice in src/run.ts
  • deploy path via GitHub Actions + Alchemy

Next:

  • hosted remote-computer runner
  • real cloud cloudbox.run() execution
  • tool plugin reference docs

File map

web/src/pages/index.astro          homepage/funnel
web/src/pages/demo.astro           demo page shell
web/src/components/SampleAgent.tsx demo client interaction
src/run.ts                         local proof-run slice
src/computer-do.ts                 workspace Durable Object
src/grade.ts                       receipt grading
alchemy.run.ts                     Cloudflare infra/deploy
scripts/demo.mjs                   headless demo
scripts/browser-demo.mjs           browser demo

Read the full file on GitHub · 125 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. 2d ago First seen · 125 lines · 670 tokens per session scan A 887b7a892b9a

Subscribe to this mod's changes

cloudbox AGENTS.md is an instructions file published in the GitHub repository acoyfellow/cloudbox (49 stars, last pushed 1mo ago), licensed MIT. It adds 670 tokens to every session, about $0.0034 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.