Platform wheels for RDKit - a comprehensive cheminformatics and machine-learning library with Python bindings
89
Pending
Does it follow best practices?
Impact
89%
1.01xAverage score across 10 eval scenarios
Pending
The risk profile of this skill
{
"context": "This evaluation assesses how well the engineer uses RDKit's BRICS fragmentation capabilities to break down molecules into chemically meaningful fragments. The focus is on correct usage of SMILES parsing, BRICS decomposition, and fragment handling functions.",
"type": "weighted_checklist",
"checklist": [
{
"name": "SMILES parsing",
"description": "Uses rdkit.Chem.MolFromSmiles() to parse the input SMILES string into a molecule object",
"max_score": 20
},
{
"name": "BRICS decomposition",
"description": "Uses rdkit.Chem.BRICS.BRICSDecompose() to fragment the molecule into chemically meaningful pieces based on synthetic disconnection points",
"max_score": 40
},
{
"name": "Fragment conversion",
"description": "Converts fragment objects to SMILES strings using rdkit.Chem.MolToSmiles() or equivalent methods to produce the output fragment list",
"max_score": 20
},
{
"name": "Fragment uniqueness",
"description": "Ensures fragments are deduplicated (e.g., using set operations) so each unique structure appears only once in the output",
"max_score": 10
},
{
"name": "Error handling",
"description": "Properly handles invalid SMILES input by checking if MolFromSmiles() returns None and providing appropriate error messages",
"max_score": 10
}
]
}evals
scenario-1
scenario-2
scenario-3
scenario-4
scenario-5
scenario-6
scenario-7
scenario-8
scenario-9
scenario-10