NeqSim is a library for calculation of fluid behavior, phase equilibrium and process simulation
Agent Claude Code
Performs root cause analysis on process equipment using NeqSim's diagnostics framework — integrates multi-source reliability data (IOGP/SINTEF, CCPS, IEEE 493, Lees, OREDA), plant historian time-series, STID design conditions, and process simulation to produce ranked failure hypotheses with Bayesian confidence scoring.
Agent Claude Code
Routes engineering requests to the most appropriate NeqSim specialist agent. Analyzes the user's question, classifies it by domain, and delegates to the right agent — or composes multi-agent pipelines for complex cross-discipline tasks. Use this when you're unsure which agent to pick.
Agent Claude Code
Performs process safety simulations — vessel depressurization/blowdown, relief valve sizing (API 520/521), trapped-liquid fire rupture screening, fire case modeling, source term generation for consequence analysis (PHAST/FLACS/KFX), safety envelope calculations (hydrate, MDMT, CO2 freezing), and risk analysis with…
Agent Claude Code
Takes a process simulation task description and delivers a complete, tested Jupyter notebook. Uses neqsimdevsetup for local development and produces Colab-compatible output. This is the fast path for getting working process simulation notebooks.
Agent Claude Code
Solves process-engineering problems using the NeqSim Java API with proportional depth: quick answer-first calculations, notebook-backed studies, or full reports. Delivers auditable task folders with validation evidence, runner-executed notebooks/scripts, and reports when needed. When NeqSim lacks a needed capability…
Agent Claude Code
Reviews NeqSim process systems, calculations, and extracted technical documents against standards and technical requirements such as DNV-ST-F101 pipeline screening, DNV-RP-F101 corroded-pipeline screening, DNV-RP-F104 CO2-pipeline envelope screening, DNV-RP-F105 free-span screening, DNV-RP-F109 on-bottom stability…
Agent Claude Code
Reads and extracts structured engineering data from technical documents and engineering images — equipment data sheets, design basis, heat & mass balance tables, technical requirements, well test reports, inspection reports, piping specifications, material certificates, standards documents, P&ID drawings, mechanical…
Agent Claude Code
Creates and configures NeqSim thermodynamic fluids and generates, plots, interprets, or troubleshoots PT phase envelopes. Selects the EOS, adds components, sets mixing rules, runs flash calculations, calculates dew/bubble curves, cricondenbar, cricondentherm, and critical points, and diagnoses Michelsen continuation…
Agent Claude Code
Reads Honeywell UniSim Design / Aspen HYSYS .usc files via COM automation and converts them to running NeqSim ProcessSystem / ProcessModule models. Extracts fluid packages, components, operations (45+ typed operation handlers including reactors, columns, controllers, adapters), streams, sub-flowsheets, and topology.…
Agent Claude Code
Coordinates screening-level utility-system DESIGN with NeqSim — sizes fired steam boilers, deaerators, vapour-compression refrigeration, on-site nitrogen generation, and multi-pressure steam networks, aggregates plant utility demands from a run flowsheet, and runs the agentic closed loop that optimizes a two-stage…
Agent Claude Code
Acts as a hostile peer reviewer for a draft paper BEFORE submission. Reads paper.md + results.json + approvedclaims.json and returns a structured critique covering novelty, evidence, statistics, replication, framing, figures, and discussion depth. Produces a list of required fixes that scientific-writer must address…
Agent Claude Code
Inspects NeqSim's current flash algorithm implementation, identifies improvement opportunities, proposes code-level changes with pseudocode, and creates implementation tickets. Does NOT merge code — all changes go through validation.
Agent Claude Code
Runs large-scale computational experiment suites using NeqSim. Generates test cases, executes baseline and modified algorithms, records convergence metrics, and produces structured result files. The truth source for all paper claims.
Agent Claude Code
Reviews PaperLab books for release readiness, focusing on pedagogy, figure evidence, source traceability, computation, citations, and readability.
Agent Claude Code
Detects repeated text, duplicated figures, repeated headings, and chapter merge candidates in large PaperLab books. Produces a concise restructuring plan and applies approved edits with quality gates enabled.
Agent Claude Code
Runs the complete PaperLab book release process, coordinates prerequisite audits, stops on blockers, renders configured outputs, and writes a release gate report.
Agent Claude Code
Finds publishable papers hidden inside PaperLab books, including benchmark suites, review sections, case studies, teaching datasets, and validated workflows.
Agent Claude Code
Tracks recurring cases across a PaperLab book and prevents assumptions, numbers, names, and teaching roles from drifting unintentionally.
Agent Claude Code
Aggregates PaperLab book audit outputs into a chapter health dashboard with release classifications and concrete next fixes.
Agent Claude Code
Adds cross-figure and cross-section synthesis paragraphs to PaperLab book chapters so visual evidence becomes a coherent teaching narrative.
Agent Claude Code
Audits and improves how key concepts recur through a PaperLab book, ensuring they are introduced, practiced, deepened, assessed, and reused coherently.
Agent Claude Code
Builds and reviews whole-book prerequisite graphs so concepts are introduced, reinforced, assessed, and cross-referenced in a coherent learning sequence.
Agent Claude Code
Curates validation datasets and source references for PaperLab papers and books, including units, provenance, uncertainty, licenses, and citation links.
Agent Claude Code
Audits equations, units, symbols, notation, and formula variants across a PaperLab book.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: