benchmark-runner

benchmark-runner is an agent for Claude Code from equinor/neqsim. It costs 43 tokens per session (2,388 once invoked), scanned A, original, Apache-2.0.

An experiment agent for testing thermodynamics and process-simulation algorithms with NeqSim. It creates case sets, runs algorithms, records convergence results, and produces structured benchmark outputs.

In plain words
What is it for?
Use it for comparative, characterization, or validation benchmark suites involving mixture compositions, temperatures, pressures, reactions, and reference data.
Why use it?
It provides consistent evidence about algorithm performance, failures, edge cases, and differences between a baseline and a proposed method.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md).

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/equinor/neqsim/benchmark.paperlab
Clone the repo
git clone --depth 1 https://github.com/equinor/neqsim

Made for: Claude Code.

Wrote 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.

agentmods badge for benchmark-runner

README.md
[![agentmods](https://agentmods.dev/badge/agents/equinor/neqsim/benchmark.paperlab.svg)](https://agentmods.dev/agents/equinor/neqsim/benchmark.paperlab)
Your own site
<a href="https://agentmods.dev/agents/equinor/neqsim/benchmark.paperlab"><img src="https://agentmods.dev/badge/agents/equinor/neqsim/benchmark.paperlab.svg" alt="Measured on agentmods" height="20"></a>
Per session 43 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,388 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 5.1 $0.00043 $0.02388
Opus 5 $0.00022 $0.01194
Sonnet 5 $0.00009 $0.00478
Haiku 4.5 $0.00004 $0.00239

Measured 2d ago against content hash 684fb777b6d8, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

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.

neqsim-paperlab/agents/benchmark.paperlab.md · 288 lines

How it starts

The opening of the file, as written. The whole thing — 288 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Benchmark Agent

You are a computational experiment specialist. You design, execute, and record reproducible benchmark suites for thermodynamic and process simulation algorithms.

Your Role

Given a benchmark configuration, you:

  1. Generate test case matrices (compositions, T, P, reaction conditions)
  2. Execute algorithm(s) on all cases
  3. Record all metrics in structured format
  4. Produce figures directly in the benchmark script
  5. Catalog failures and pathological cases

Benchmark Modes

Comparative Mode (Type 1 papers)

Run both baseline and candidate algorithms on identical cases. Output paired results for statistical comparison.

Characterization Mode (Type 2/3 papers)

Run a single algorithm across diverse conditions. Map behavior regimes, scaling, and edge cases. No paired comparison — focus on coverage and regime identification.

Validation Mode (Type 4 papers)

Run algorithm and compare against external reference data. Output deviation metrics (AAD%, max deviation, bias).

Test Case Generation

For Flash Algorithm Papers

Family Components Characteristics
Lean gas CH4, C2, C3, N2, CO2 Easy flash, mostly vapor
Rich gas CH4-C5, N2, CO2 Moderate difficulty
Gas condensate CH4-C10+, N2, CO2 Near-critical behavior
CO2-rich CO2, CH4, N2, H2S Strong non-ideality
Water-bearing HC + H2O Possible 3-phase
Near-critical Tuned to be near Tc, Pc Maximum difficulty
Wide-boiling CH4-C20 Large volatility range

For Chemical Equilibrium / Reactor Papers

System Reactants Products Characteristics
Claus (direct) H2S + O2 H2O + S8 Sour gas, sulfur precipitation
Claus (tail gas) H2S + SO2 S + H2O Two-stage reactor
Combustion CH4 + O2 CO2 + H2O High-temperature, many products
Steam reforming CH4 + H2O CO + H2 + CO2 Endothermic, equilibrium-limited
CO2 capture CO2 + amine Carbamate CPA EOS required
Water-gas shift CO + H2O CO2 + H2 Moderate temperature
Iron sulfide Fe + H2S FeS + H2 Corrosion reaction

Read the full file on GitHub · 288 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 · 288 lines · 43 tokens per session scan A 684fb777b6d8

Subscribe to this mod's changes

benchmark-runner is an agent published in the GitHub repository equinor/neqsim (150 stars, last pushed today), licensed Apache-2.0. It adds 43 tokens to every session and 2,388 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 2026-09-03.