Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Query PubChem, search by name/CID/SMILES, retrieve properties, similarity/substructure searches, bioactivity, for cheminformatics. Use when a user asks about a specific chemical, drug, or molecule.
.claude/skills/mkurman-pubchem-database/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-22 | ✓→✗ | ▼ Worse | 124% | 0% |
| case-20 | ✓→✓ | = Same ✓ | 136% | 0% |
| case-21 | ✓→✓ | = Same ✓ | 174% | 0% |
| case-16 | ✗→✗ | = Same ✗ | 726% | 0% |
| case-17 | ✗→✗ | = Same ✗ | 442% | 0% |
uv: Read the uv skill and follow its Setup instructions to ensureuv is installed and on PATH.
this skill directory then (1) prominently notify the user to check the terms at https://pubchem.ncbi.nlm.nih.gov/docs/citation-guidelines and https://pubchem.ncbi.nlm.nih.gov/docs/pug-rest, then (2) create the file recording the notification text and timestamp.
PubChem queries are executed via a robust Python wrapper script to respect terms-of-service and handle complex JSON parsing. This script allows safe multi-agent use of APIs.
Example: Resolve a chemical name to its Compound ID (CID)
bashuv run scripts/pubchem_api.py resolve --name "aspirin" --output result.json
database rather than accessing the database directly. The scripts automatically enforce the required rate limit gracefully.
database query if the user asks for a specific fact that can be checked in PubChem. Do not rely solely on internal knowledge.
output.
1. Compound Resolution (Name or InChI to Identifiers) Convert chemical/trade names or InChI strings into PubChem CIDs, SMILES, and InChIKeys.
bashuv run scripts/pubchem_api.py resolve --name "ibuprofen" --output result.json # OR uv run scripts/pubchem_api.py resolve --inchi "InChI=1S/C3/c1-3-2/i1+1" --output result.json
2. Physical & Chemical Property Retrieval Fetch computed properties (e.g., MolecularWeight, XLogP, TPSA).
bashuv run scripts/pubchem_api.py properties --cid 2244 --output result.json
3. Synonyms and Trade Names Find alternative names and brand names.
bashuv run scripts/pubchem_api.py synonyms --cid 2244 --output result.json
4. Safety and Hazard Information (GHS) Retrieve Global Harmonized System hazard statements and handling precautions (uses PUG-View).
bashuv run scripts/pubchem_api.py safety --cid 2244 --output result.json
5. Drug and Medication Information Fetch FDA pharmacology data, mechanism of action, and therapeutic uses (uses PUG-View).
bashuv run scripts/pubchem_api.py pharmacology --cid 2244 --output result.json
6. Custom Heading (PUG-View) Retrieve any specific heading from the PUG-View system (e.g., 'Geometry', 'Crystal Structures').
bashuv run scripts/pubchem_api.py view --cid 3939 --heading "Crystal Structures" --output result.json
7. Image Generation Retrieve 2D chemical structure images. The script returns a Markdown-formatted image link.
bashuv run scripts/pubchem_api.py image --cid 2244 --output result.json
8. Structure-Based Searching (Similarity & Substructure) Find molecules similar to a SMILES string or containing a specific substructure.
bashuv run scripts/pubchem_api.py similarity --smiles "CC(=O)OC1=CC=CC=C1C(=O)O" --output result.json
and
bashuv run scripts/pubchem_api.py substructure --smiles "C1=CC=CC=C1" --output result.json
9. BioAssay & Target Interactions Identify genes or proteins a chemical interacts with.
bashuv run scripts/pubchem_api.py assays --cid 2244 --output result.json
10. Cross-references (Xrefs) Fetch identifiers cross-referenced with a CID (e.g., PatentID, PubMedID).
bashuv run scripts/pubchem_api.py xrefs --cid 2244 --type "PatentID" --output result.json
11. Property Range Search Find CIDs within a specific property range. Supported features include: molecular_weight, heavy_atom_count, xlogp, tpsa, h_bond_donor_count, h_bond_acceptor_count, rotatable_bond_count, exact_mass, monoisotopic_mass, and complexity.
bashuv run scripts/pubchem_api.py range --feature molecular_weight --min 400.0 --max 400.05 --output result.json
12. Custom PUG-REST Query Execute a raw path against the PUG-REST API.
bashuv run scripts/pubchem_api.py query --path "compound/cid/2244/xrefs/PatentID/JSON" --output result.json
If direct resolution by name or formula fails (e.g., for complex compounds or specific ions):
searching for the neutral parent compound.
try searching for major components or ligands.
can generate one for a component, use it to find related compounds.
references/endpoints.md to construct raw PUG-REST URLs.
follow the checklists in references/workflows.md.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-16 | fail→fail | 35,335 | 7,755 | -78% | 1 | 1 | 0% | 234 | 1,932 | +726% | 0 | 0 | — |
case-17 | fail→fail | 8,692 | 6,506 | -25% | 1 | 1 | 0% | 316 | 1,714 | +442% | 0 | 0 | — |
case-18 | fail→fail | 28,534 | 5,715 | -80% | 1 | 1 | 0% | 1,222 | 1,615 | +32% | 0 | 0 | — |
case-01 | fail→fail | 7,081 | 4,945 | -30% | 1 | 1 | 0% | 1,262 | 1,745 | +38% | 0 | 0 | — |
case-02 | fail→fail | 10,999 | 6,962 | -37% | 1 | 1 | 0% | 1,974 | 1,914 | -3% | 0 | 0 | — |
case-03 | fail→fail | 29,030 | 7,353 | -75% | 1 | 1 | 0% | 6,203 | 1,946 | -69% | 0 | 0 | — |
case-04 | fail→fail | 7,663 | 6,046 | -21% | 1 | 1 | 0% | 1,574 | 1,831 | +16% | 0 | 0 | — |
case-05 | fail→fail | 30,725 | 6,949 | -77% | 1 | 1 | 0% | 6,093 | 1,874 | -69% | 0 | 0 | — |
case-06 | fail→fail | 13,861 | 7,053 | -49% | 1 | 1 | 0% | 2,292 | 1,653 | -28% | 0 | 0 | — |
case-07 | fail→fail | 20,775 | 5,364 | -74% | 1 | 1 | 0% | 3,871 | 1,683 | -57% | 0 | 0 | — |
case-08 | fail→fail | 6,357 | 5,309 | -16% | 1 | 1 | 0% | 298 | 1,699 | +470% | 0 | 0 | — |
case-09 | fail→fail | 14,362 | 6,692 | -53% | 1 | 1 | 0% | 2,025 | 1,809 | -11% | 0 | 0 | — |
case-10 | fail→fail | 12,606 | 5,982 | -53% | 1 | 1 | 0% | 2,113 | 1,700 | -20% | 0 | 0 | — |
case-11 | fail→fail | 12,499 | 6,058 | -52% | 1 | 1 | 0% | 1,569 | 1,716 | +9% | 0 | 0 | — |
case-12 | fail→fail | 8,020 | 6,435 | -20% | 1 | 1 | 0% | 1,590 | 1,759 | +11% | 0 | 0 | — |
case-13 | fail→fail | 10,165 | 6,213 | -39% | 1 | 1 | 0% | 1,728 | 1,914 | +11% | 0 | 0 | — |
case-14 | fail→fail | 6,624 | 7,444 | +12% | 1 | 1 | 0% | 218 | 1,867 | +756% | 0 | 0 | — |
case-15 | fail→fail | 7,420 | 4,732 | -36% | 1 | 1 | 0% | 886 | 1,609 | +82% | 0 | 0 | — |
case-19 | fail→fail | 9,367 | 4,649 | -50% | 1 | 1 | 0% | 1,382 | 2,306 | +67% | 0 | 0 | — |
case-20 | pass→pass | 5,733 | 6,203 | +8% | 1 | 1 | 0% | 1,112 | 2,621 | +136% | 0 | 0 | — |
case-21 | pass→pass | 6,691 | 11,409 | +71% | 1 | 1 | 0% | 1,357 | 3,720 | +174% | 0 | 0 | — |
case-22 | pass→fail | 9,019 | 17,104 | +90% | 1 | 1 | 0% | 1,674 | 3,751 | +124% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 3 counted toward the lift figure. The other 19 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of -5 percentage points is the difference between those two pass rates over the 3 comparable cases. 3 cases got worse with the skill loaded, and they are included in that figure.
Other measured skills in the registry, with their headline benchmark lift.