Medicinal Chemistry Compound Filters
Screens drug candidate molecules with drug-likeness rules, structural alerts and complexity filters.
- Repository
- K-Dense-AI/scientific-agent-skills
- Stars
- 55,780
- Licence
- MIT
- Files
- 4 files
What it does
Lets an AI agent triage large lists of chemical compounds the way a medicinal chemist would, by applying published drug-likeness rules (such as Lipinski and Veber), structural warning-sign catalogs (such as PAINS and NIBR), and complexity thresholds. It uses the open-source medchem Python library to do the calculations and returns tables of which molecules pass or fail each filter. The skill itself is guidance plus a batch-filtering script; it does not replace expert judgement about a specific drug target.
Great for
- Medicinal chemists or drug discovery teams narrowing a compound library before lab testing
- Data scientists building reproducible molecular filtering pipelines
- Students learning how drug-likeness rules and structural alerts are applied
Skip it if
- Clinical or safety decisions about real medicines: the filters are screening heuristics, not toxicity or efficacy tests, and the skill itself warns they are context-specific
- Users without Python, RDKit and a working chemistry environment, since setup is technical and the optional Lilly tools need a C++ compiler, Ruby and sometimes WSL
Try asking
- “Filter this CSV of compounds with Lipinski and Veber rules and PAINS alerts, and save the passing molecules”
- “Check these SMILES strings for CNS-likeness and report which ones fail each rule”
- “Run NIBR screening-deck filters on my SDF library and flag covalent warheads”
- “Write a query that keeps molecules matching rule_of_five and excluding PAINS alerts”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 4 original files, the MIT licence and an attribution note.
medchem.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip medchem.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip medchem.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Validate that input SMILES parse correctly before filtering, since some functional filters can admit parse failures.
- Pass copies of molecules to the bredt_filter, as its in-place kekulization changes later aromatic alert results.
- Treat filter results as a first-pass triage and review flagged compounds with a domain expert.
What's inside
- references/api_guide.md
- references/rules_catalog.md
- scripts/filter_molecules.py
- SKILL.md
Redistributed unmodified from K-Dense-AI/scientific-agent-skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.