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.
npx agentmods add skills/chainbench/openchainbench/contribute-benchmarknpx skills add ChainBench/OpenChainBench --skill contribute-benchmarkgit clone --depth 1 https://github.com/ChainBench/OpenChainBenchWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/chainbench/openchainbench/contribute-benchmark)<a href="https://agentmods.dev/skills/chainbench/openchainbench/contribute-benchmark"><img src="https://agentmods.dev/badge/skills/chainbench/openchainbench/contribute-benchmark.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00039 | $0.01969 |
| Opus 5 | $0.00019 | $0.00984 |
| Sonnet 5 | $0.00008 | $0.00394 |
| Haiku 4.5 | $0.00004 | $0.00197 |
Grade A, and why
openchainbench-contributor 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are helping someone add a new benchmark to OpenChainBench, an open registry of crypto-infrastructure benchmarks.
If the user does not already have the repo cloned, your first step is to suggest git clone https://github.com/ChainBench/OpenChainBench && cd OpenChainBench. Everything below assumes that working directory.
Mental model
OpenChainBench is a federation:
- Editorial copy and queries live in
benchmarks/<slug>.yml(one YAML per benchmark). - A harness produced by the contributor exposes
/metricsover HTTPS at a stable public URL. - A single shared Prometheus scrapes every harness on a schedule.
- The Next.js site at
openchainbench.comreads from that Prometheus through the ISR cycle (revalidates every 60 seconds).
The contributor owns the harness runtime, the secrets and the budget. Maintainers never see the contributor's API keys.
Workflow
Walk the user through these six steps in order. Do not skip ahead.
1. Open an issue
Direct them to the Propose a benchmark issue template at https://github.com/ChainBench/OpenChainBench/issues/new?template=new-benchmark.yml. Maintainers need to see the proposed metric, providers, methodology and harness location before code review. Brainstorm-stage ideas belong in https://github.com/ChainBench/OpenChainBench/discussions/categories/ideas instead.
2. Write the spec
Path: benchmarks/<slug>.yml. The Zod schema in src/lib/spec-schema.ts is the authoritative reference. Required fields and their semantics:
| Field | What |
|---|---|
slug |
Kebab-case identifier matching the filename. Used in URLs. |
number |
Zero-padded benchmark number ("009", "010"...). Read the highest existing number in benchmarks/ and increment. |
title |
Short display name. Goes into the page H1. Stay factual, no marketing words. |
seo_title |
Optional longer title used in <title> and OG, can mention competitor names for long-tail search. |
subtitle |
One-line summary of what is measured. |
category |
One of Aggregators, Bridges, Blockchains, Trading, Wallets, RPCs. |
metric |
The thing being measured (Head lag, Quote latency, All-in fee, ...). |
unit |
One of ms, s, pct, bps, count. The site formats numbers accordingly. |
higher_is_better |
false for latency or fees, true for coverage / uptime. |
abstract |
Multi-paragraph description shown under the H1. |
methodology |
Array of bullet points. Routes, notional sizes, cadence, region, exclusion criteria. |
source |
URL to the harness source on GitHub. Required even for drafts. |
prometheus.url |
The shared Prometheus base URL. Default: https://prom.openchainbench.com. |
prometheus.window |
Aggregation window, typically 24h. |
providers |
Array of { name, slug, url?, type?, queries: { p50, p90, p99, mean?, success?, sample_size?, series } }. |
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.
- 4d ago First seen · 146 lines · 39 tokens per session scan A 96fa07e5e357
openchainbench-contributor is a skill published in the GitHub repository ChainBench/OpenChainBench (7 stars, last pushed today), licensed MIT. It adds 39 tokens to every session and 1,969 once invoked, about $0.0002 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.
Other skills, from other repositories
tests
Write clear, maintainable Vitest and Playwright tests with precise assertions, consistent structure, and strong behavioral coverage.
defuddle
Plan and, with explicit network consent, use an optional external Defuddle cleaner to extract article-like HTTPS pages as Markdown. Use for defuddle, clean this URL, strip page clutter, readable Markdown from a web page, or preparing a web source for later wiki ingestion.
browser-automation
Playwright-based browser automation patterns for autonomous web interaction.
amazon
Shop on Amazon and Amazon Fresh through your browser.
gmgn-contract-dd
Contract due-diligence score for one token address — contract safety, holder structure and price action combined into a single 0-100 composite, capped by GMGN's own rug label, where every deduction names the field it read and an absent field is never a passing check. Use when the user wants one verdict number rather…
gmgn-market
Get crypto and meme token price charts (K-line, candlestick, OHLCV), trending meme coin rankings by volume, newly launched tokens on launchpads (pump.fun, fourmeme, letsbonk, Raydium, etc.), the hot-search ranking (most-searched tokens), and search for a specific token or wallet by name, symbol, contract address…