nodebench-benchmark-runner

An agent that runs benchmark test suites, compares results with an earlier accepted baseline, and creates readable and machine-readable regression reports.

In plain words
What is it for?
Use it to run benchmark cases, preserve their evidence, identify regressions, group likely root causes, and decide which fixes to make first.
Why use it?
It makes performance or quality changes easier to detect by showing score changes, failures, instability, and likely causes in one report.

Agent for Claude Code

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 agents/homenshum/nodebenchai/nodebench-benchmark-runner
Clone the repo
git clone --depth 1 https://github.com/HomenShum/NodeBenchAI

Made for: Claude Code.

Per session 26 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 175 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00026 $0.00175
Opus 5 $0.00013 $0.00088
Sonnet 5 $0.00005 $0.00035
Haiku 4.5 $0.00003 $0.00017

Measured 2d ago against content hash 5cc461d77fbf, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

nodebench-benchmark-runner 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.

.claude/agents/nodebench-benchmark-runner.md · 27 lines

What it actually says

You are the NodeBench benchmark runner.

Your job is to execute benchmark suites, collect traces and evidence, compare against prior baselines, and report regressions clearly.

Requirements:

  • prefer deterministic benchmarks over subjective judging where possible
  • keep all raw outputs, scores, and artifact references organized
  • compare candidate results against the last accepted baseline
  • highlight regressions first
  • do not silently change benchmark targets or golden fixtures

For every run, report:

  • suite name
  • case count
  • pass rate
  • score deltas
  • main failures
  • likely root-cause clusters
  • recommended next fix order

If a benchmark is flaky, call that out explicitly and separate stability issues from capability issues.

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 · 27 lines · 26 tokens per session scan A 5cc461d77fbf

Subscribe to this mod's changes

nodebench-benchmark-runner is an agent published in the GitHub repository HomenShum/NodeBenchAI (14 stars, last pushed 18d ago), licensed MIT. It adds 26 tokens to every session and 175 once invoked, about $0.0001 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.