What problem does it solve?
Organic chemistry problems are often answered by pattern-matching named reactions or memorized shift tables, which produces wrong products, incorrect stereochemistry, and unverified SMILES. This Skill replaces memorization with first-principles reasoning (electron flow, driving forces, kinetic vs thermodynamic control) and enforces computational verification of every numerical or structural answer.
Core Features & Use Cases
- Reaction Product Prediction: A five-step reasoning process covering reactive site identification, electron flow, driving forces, and sanity checks, plus a named-reaction decision tree (Kumada, Wittig, Diels-Alder, Pummerer, [3,3]-sigmatropic) and retrosynthetic disconnection heuristics.
- Spectroscopy Interpretation: A systematic strategy combining degrees of unsaturation, IR functional-group triage, 1H NMR shift/multiplicity/integration analysis, and 13C symmetry counting to determine structures from combined data.
- Stereochemistry Tracking: CIP priority assignment, R/S determination, and mechanism-based outcome prediction (SN2 inversion, SN1 racemization, E2 anti-periplanar, Felkin-Anh), with a script that tracks configuration through multi-step sequences.
- Verification Scripts: Bundled Python tools for SMILES property verification (MW, heavy atoms, valence electrons), combustion analysis, degrees of unsaturation, crystal density validation, and a lookup database for allotropes, point groups, and reagent selectivity.
- Use Case: Given a combustion analysis (0.2 g sample yielding 0.4874 g CO2 and 0.1998 g H2O, molar mass 78), run molecular_formula.py to derive C6H6, confirm DoU = 4, then reason through its bromination behavior.
Quick Start
Ask the AI to predict the product and mechanism of an organic reaction, for example: predict the product of excess methylmagnesium bromide with 1,3-dichlorobenzene and a palladium catalyst, and verify the answer with the bundled scripts.