Skip to content
Tresca

Chemical engineering skills (化工设计专家工作库)

PluginCommunity

A Chinese-language Codex plugin of 19 Aspen Plus design skills, with equipment sizing and optional MCP. By milklong.

9 starsUpdated 12 days agoGitHub data from 5 Nov 2026

At a glance

Type
Plugin
Works with
Aspen Plus
How it connects
Drives a running copy of the tool
Access
Can modify models
Tool versions
Not stated for the skills; the bundled copy of the ddtlxc001 Aspen MCP targets Aspen Plus V15
Operating systems
Windows
Needs a Aspen Plus licence
Yes
Transport
Local (stdio)
AI clients documented
Codex
Licence
MIT
Language
Python

What it can do

  • Takes a design brief through planning, a simplified RStoic/SEP skeleton, block-by-block rigorous modelling and a whole-plant re-run, recording sources and changes at each step.
  • Separate skills cover distillation set-up from DSTWU to RadFrac, heat-pump distillation, pressure paths, flowsheet error repair, kinetics documentation and component templates.
  • An equipment backend sizes 17 equipment types and feeds real limits, such as exchanger or compressor staging, back into the model.
  • Further skills cover Aspen EDR rating, AEPA equipment cost extraction and SW6 field mapping.
  • A local stdio MCP server offers the knowledge search and equipment calculations, and a vendored copy of the ddtlxc001 aspen-mcp server can be started separately.

Limitations

  • Written in Chinese, and aimed at Codex.
  • Aspen operations need a local, licensed Aspen Plus on Windows; EDR and SW6 skills need those products too.
  • The offline runtime was checked on CPython 3.14 for Windows AMD64; third-party reference data in the knowledge base keeps its own licences rather than MIT.

Install

python tools/verify_release.py --root .
python tools/install.py --workspace-root D:/ChemicalWork --skills-root "$env:USERPROFILE/.agents/skills"

Source: github.com/milklong888/chemical-engineering-skills/blob/main/docs/使用手册.md

MCP servers and skills run code on your computer. Review the source before installing.

AI-generated models, drawings and results must be checked by a qualified engineer before use.