ClassyFire is an automated chemical classification tool that uses a comprehensive, computable chemical taxonomy (ChemOnt) to classify all known chemical compounds based on their structural features. The tool automatically assigns compounds to one of over 4800 categories, organized into 11 hierarchical levels, with each category defined by computable structural rules. ClassyFire's webserver is freely accessible at http://classyfire.wishartlab.com/, and a Ruby API is available at https://bitbucket.org/wishartlab/classyfire_api. ClassyFire has been used to annotate over 77 million compounds and has been integrated into other software packages to generate textual descriptions and infer biological properties of over 100,000 compounds. ClassyFire's classification process involves four steps: (1) creation and preprocessing of the chemical entity, (2) feature extraction, (3) rule-based category assignment and reduction, and (4) selection of the direct parent. The tool uses a combination of structural features, logical expressions, and chemical formulae to classify compounds. It also provides a molecular framework and a list of chemical substituents, along with a text-based description of the compound. ClassyFire has been tested on a set of 800 compounds not used during its training phase, and its classification results have been validated against expert annotations. ClassyFire's taxonomy has been mapped to other chemical ontologies, including ChEBI, LIPID MAPS, and MeSH, allowing for cross-referencing with other chemical databases. The tool is also capable of classifying chemical mixtures and has been used to classify over 70 million compounds. ClassyFire's results have been used to annotate compounds in databases such as ChEBI, LIPID MAPS, and PubChem. The tool is freely available and provides a powerful text-based search interface, making it a valuable resource for chemists and cheminformatics researchers.ClassyFire is an automated chemical classification tool that uses a comprehensive, computable chemical taxonomy (ChemOnt) to classify all known chemical compounds based on their structural features. The tool automatically assigns compounds to one of over 4800 categories, organized into 11 hierarchical levels, with each category defined by computable structural rules. ClassyFire's webserver is freely accessible at http://classyfire.wishartlab.com/, and a Ruby API is available at https://bitbucket.org/wishartlab/classyfire_api. ClassyFire has been used to annotate over 77 million compounds and has been integrated into other software packages to generate textual descriptions and infer biological properties of over 100,000 compounds. ClassyFire's classification process involves four steps: (1) creation and preprocessing of the chemical entity, (2) feature extraction, (3) rule-based category assignment and reduction, and (4) selection of the direct parent. The tool uses a combination of structural features, logical expressions, and chemical formulae to classify compounds. It also provides a molecular framework and a list of chemical substituents, along with a text-based description of the compound. ClassyFire has been tested on a set of 800 compounds not used during its training phase, and its classification results have been validated against expert annotations. ClassyFire's taxonomy has been mapped to other chemical ontologies, including ChEBI, LIPID MAPS, and MeSH, allowing for cross-referencing with other chemical databases. The tool is also capable of classifying chemical mixtures and has been used to classify over 70 million compounds. ClassyFire's results have been used to annotate compounds in databases such as ChEBI, LIPID MAPS, and PubChem. The tool is freely available and provides a powerful text-based search interface, making it a valuable resource for chemists and cheminformatics researchers.