Your verified CalcTree pages are now callable tools for any AI. Instead of trusting an LLM to do engineering maths, it calls your calculation page, overrides the inputs, and reads back results with real physical units.
Download the skill · Setup docs
How it works
The CalcTree LLM Skill teaches any AI assistant three things: how to discover pages in your workspace, how to execute any page with custom inputs, and how to read back typed results with units.
The AI does the reasoning and interpretation. CalcTree's engine does the arithmetic. The numbers are deterministic, unit-aware, and come from a page your team can read, audit, and version.
Why it matters
LLMs are good at reasoning but unreliable at arithmetic. Engineering calculations need both. By giving the AI access to CalcTree as a tool, you get the LLM's ability to interpret problems, choose inputs, and explain results, backed by a calculation engine that carries real physical units through every step.
Ask it to check a steel beam with a 6 m span and a 16.5 kN/m load. It calls your beam calculator, gets back Mmax = 74.25 kNm and a deflection of 15.4 mm against a 24 mm limit, and reports the section is adequate at 67% utilisation. No hallucinated numbers.
What else it can do
The skill also lets your AI build new calculation pages from scratch and export any page as a formatted PDF report with a title page, table of contents, and math notation.
Get started
You need an API key and the skill file. Generate a key in your workspace settings, download the skill from GitHub, and follow the setup steps for your AI tool in the docs.
Works with Claude, ChatGPT, Gemini, Cursor, and any LLM that supports skills or tool use.
