chem-ts-optimization

Optimize non-periodic molecular transition-state guesses and validate first-order saddle character via vibrational analysis.

144|21|Updated Jan 8, 2026
One-click install
npx skills add https://github.com/learningmatter-mit/AtomisticSkills --skill chem-ts-optimization
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: chem-ts-optimization
Source: https://github.com/learningmatter-mit/AtomisticSkills/tree/main/.agents/skills/chem-ts-optimization
Command: npx skills add https://github.com/learningmatter-mit/AtomisticSkills --skill chem-ts-optimization

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill optimizes a molecular transition-state (TS) geometry from an initial guess and validates that it is a first-order saddle point using vibrational analysis.

Core Features & Use Cases

  • TS optimization for non-periodic molecules: Runs Sella-based TS optimization followed by a finite-difference vibrational frequency check.
  • First-order saddle validation: Classifies the optimized structure by counting imaginary modes below a configurable cm^-1 cutoff.
  • Backend-ready execution: Supports MLIP backends for scoring/optimization (MACE or FAIRChem/UMA) and writes a complete results bundle.

Quick Start

Use the chem-ts-optimization Skill to optimize a TS guess for a non-periodic molecule and confirm first-order saddle character by generating ts_optimized.xyz and ts_optimization_results.json from a provided ts_guess.xyz file.

Frequently Asked Questions about chem-ts-optimization

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I optimize a molecular transition-state geometry and verify a first-order saddle point?

You can optimize a molecular transition-state geometry by providing an initial guess structure for Sella-based TS optimization, followed by a finite-difference vibrational analysis to confirm it is a first-order saddle point with exactly one imaginary mode below a cutoff.

What is a first-order saddle point and how does vibrational analysis confirm it?

A first-order saddle point is a transition-state geometry with exactly one imaginary vibrational mode. Vibrational analysis confirms this by counting imaginary frequencies below a cm^-1 cutoff to validate the optimized structure.

Can I use MLIP backends like MACE or FAIRChem for transition-state optimization?

Yes, MLIP backends like MACE or FAIRChem/UMA are supported for scoring and optimizing non-periodic molecular transition-state geometries during the Sella optimization process.

Does this transition-state optimization workflow support periodic structures?

No, this transition-state optimization workflow does not support periodic structures. It explicitly enforces pbc=False, restricting the TS optimization and vibrational analysis to non-periodic molecular systems.

What output files are generated after running Sella TS optimization and vibrational analysis?

The workflow generates ts_optimized.xyz containing the final geometry and ts_optimization_results.json with structured outputs including n_imag_below_cutoff and is_first_order_saddle.