What problem does it solve?
This Skill removes manual lookups by automating PubChem Compound queries to retrieve standardized identifiers, computed properties, and optional structure files for downstream drug-discovery or cheminformatics workflows.
Core Features & Use Cases
- Multi-modal compound search: Find PubChem CIDs by name, SMILES, InChI, InChIKey, CID, or molecular formula.
- Property retrieval for automation: Pull key computed properties (e.g., molecular weight, XLogP, TPSA, HBD/HBA, rotatable bonds, canonical/isomeric SMILES, InChI/InChIKey, IUPAC name) in a reproducible JSON output.
- Optional enrichment: Retrieve a limited set of synonyms and optionally download 2D/3D SDF structures with 3D-to-2D fallback when available.
- Rate-limited, retry-aware querying: Uses adaptive rate limiting and exponential backoff to handle PubChem throttling (503) safely for batch/agent use.
Quick Start
Query PubChem for aspirin by name (exact match) and write the results to a JSON file: python .agents/skills/drug-db-pubchem/scripts/query_pubchem.py --name "aspirin" --name_type complete --max_results 5 --outdir research/pubchem/aspirin --output aspirin.json