What is OpenChainBench AGENTS.md?

Project instructions for OpenChainBench, a Next.js website with separate staging and production branches.

Use them when changing code, opening pull requests, or deploying OpenChainBench. They explain which guides to read, where changes should merge, and how staging previews work.

They prevent outdated Next.js assumptions and reduce the risk of deploying unfinished changes to the live site.

OpenChainBench AGENTS.md — Instructions file by ChainBench

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/chainbench/openchainbench/agents-md

Clone the repo

git clone --depth 1 https://github.com/ChainBench/OpenChainBench

Made for: Codex, OpenCode.

Per session 467 This file is loaded in full into every session.

When invoked 467 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	$0.00467	$0.00467
Opus 5	$0.00234	$0.00234
Sonnet 5	$0.00093	$0.00093
Haiku 4.5	$0.00047	$0.00047

Measured 2d ago against content hash 4cbdb914b17a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

OpenChainBench 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 · 27 lines

What it actually says

This is NOT the Next.js you know

This version has breaking changes. APIs, conventions, and file structure may all differ from your training data. Read the relevant guide in node_modules/next/dist/docs/ before writing any code. Heed deprecation notices.

Branch workflow

Two long-lived branches:

main → openchainbench.com (production). Every push to main triggers .github/workflows/prod-deploy.yml which runs vercel --prod automatically via CI.
dev → staging. Each push triggers .github/workflows/staging-deploy.yml which calls Vercel CLI with a token and emits a Preview URL.

Standard flow:

Open a feature branch from dev: git checkout dev && git pull && git checkout -b feat/your-thing
PR feat/your-thing → dev (NOT main). After merge, the staging Action posts a Preview URL in the workflow summary.
Review on the Preview URL.
When the staging state looks ready to ship, open a PR dev → main. After merge, prod-deploy.yml runs automatically.

Rules:

Never push directly to main or dev. PRs only.
The staging URL emits <meta robots=noindex> and serves robots.txt: Disallow /, see src/app/robots.ts and src/app/layout.tsx. Don't undo without thinking through duplicate-content SEO impact on the prod domain. Vercel Preview SSO is disabled (URL is publicly shareable), so the noindex is the only thing keeping crawlers out.
Hotfix path: if prod is on fire and waiting through dev is unacceptable, open the fix PR directly against main, merge (CI deploys automatically), then immediately fast-forward dev from main (git checkout dev && git merge --ff-only main && git push) so the two branches don't diverge.
Required GitHub secrets (repo-level, not environment-scoped): VERCEL_TOKEN, VERCEL_ORG_ID, VERCEL_PROJECT_ID.

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 · 27 lines · 467 tokens per session scan A 4cbdb914b17a

Subscribe to this mod's changes

OpenChainBench AGENTS.md is an instructions file published in the GitHub repository ChainBench/OpenChainBench (6 stars, last pushed 2d ago), licensed MIT. It adds 467 tokens to every session, about $0.0023 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.

api benchmark crypto data open-source reproductible

OpenChainBench AGENTS.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

What it actually says

This is NOT the Next.js you know

Branch workflow

What this file has done since we first saw it

Other instructions, from other repositories

gmgn-skills CLAUDE.md

cryptocurrency.cv AGENTS.md

cryptocurrency.cv CLAUDE.md

urantia-dev-api CLAUDE.md

Binance-MCP AGENTS.md

Binance-MCP copilot-instructions.md