optimize neqsim process

A process-optimisation add-on for NeqSim, a tool that simulates industrial processes. It adjusts selected process settings to improve a chosen result while respecting limits.

In plain words
What is it for?
Use it to optimise process flowsheets, test ranges of input values, compare trade-offs, and run Monte Carlo simulations, which repeatedly test varied inputs to study uncertainty.
Why use it?
It removes much of the manual trial and error involved in finding better operating conditions. It supports different goals, including throughput, energy use, constraints, multiple competing goals, parameter studies, and uncertainty testing.

Agent

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/optimize
Clone the repo
git clone --depth 1 https://github.com/equinor/neqsim
Per session 107 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,249 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 $0.00107 $0.01249
Opus 5 $0.00053 $0.00624
Sonnet 5 $0.00021 $0.00250
Haiku 4.5 $0.00011 $0.00125

Measured 2d ago against content hash 10e9e86e85b9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

optimize neqsim process 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.

.github/agents/optimize.agent.md · 88 lines

How it starts

The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Loaded skills: neqsim-optimization-and-doe, neqsim-agentic-process-optimization, neqsim-api-patterns, neqsim-heat-integration, neqsim-java8-rules, neqsim-notebook-patterns, neqsim-professional-reporting

Skills to Load

ALWAYS read these skills before proceeding:

  • .github/skills/neqsim-optimization-and-doe/SKILL.md — decision tree, all 30 optimizer classes, code patterns
  • .github/skills/neqsim-api-patterns/SKILL.md — fluid creation, flash, equipment
  • .github/skills/neqsim-heat-integration/SKILL.md — when objective involves utility cost or energy efficiency
  • .github/skills/neqsim-java8-rules/SKILL.md — Java 8 compatibility
  • .github/skills/neqsim-notebook-patterns/SKILL.md — when delivering notebook output
  • .github/skills/neqsim-professional-reporting/SKILL.md — when delivering a task report

Operating Principles

  1. Never reinvent. NeqSim has ~30 optimizer classes. Use the decision tree in the skill.
  2. Classify first: throughput / custom-objective / constrained-NLP / Pareto / DoE / Monte Carlo.
  3. Set bounds always. Every decision variable needs physically meaningful [lo, hi].
  4. Run base case first. Always process.run() before invoking any optimizer.
  5. Check convergence. Inspect OptimizationResult.isConverged() and getConstraintViolations().
  6. Validate with sensitivity around the optimum — local sensitivity + at least one Monte Carlo screening of the most uncertain inputs.
  7. Reproducibility. Save the YAML spec via ProductionOptimizationSpecLoader whenever practical.
  8. Honest about gaps. Do NOT claim Bayesian optimization, MINLP, or LHS exist in NeqSim today — escalate to external tools via ProcessSimulationEvaluator instead.

Algorithm Selection Quick Reference

Problem Class + algorithm
Max throughput, monotonic feasibility ProcessOptimizationEngine + BINARY_SEARCH
Max throughput, non-monotonic ProcessOptimizationEngine + GOLDEN_SECTION
2–10 vars, smooth, custom objective ProductionOptimizer + NELDER_MEAD_SCORE
Non-convex, multi-modal ProductionOptimizer + PARTICLE_SWARM_SCORE
Constrained NLP (eq + ineq + bounds) SQPoptimizer
Pareto, 2–4 objectives MultiObjectiveOptimizer.optimizeWeightedSum
Parameter sweep / DoE BatchStudy.builder(...).vary(...).parallelism(n)
Uncertainty + tornado MonteCarloSimulator
Auto-size + optimize DesignOptimizer.forProcess(p)...optimize()
External SciPy / Pyomo / BoTorch ProcessSimulationEvaluator

Read the full file on GitHub · 88 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 · 88 lines · 107 tokens per session scan A 10e9e86e85b9

Subscribe to this mod's changes

optimize neqsim process is an agent published in the GitHub repository equinor/neqsim (147 stars, last pushed 2d ago), licensed Apache-2.0. It adds 107 tokens to every session and 1,249 once invoked, about $0.0005 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.