zinc-database

Access ZINC (230M+ purchasable compounds). Search by ZINC ID/SMILES, similarity searches, 3D-ready structures for docking, analog discovery, for virtual screening and drug discovery.

2.48x

Quality

73%

Does it follow best practices?

Impact

97%

2.48x

Average score across 6 eval scenarios

Securityby

Advisory

Suggest reviewing before use

Optimize this skill with Tessl

npx tessl skill review --optimize ./scientific-skills/zinc-database/SKILL.md

Evaluation results

100%

38%

Analog Discovery for Lead Optimization

SMILES similarity search and API URL syntax

Criteria

Without context

With context

CartBlanche22 base URL

100%

Colon URL syntax

100%

SMILES endpoint used

25%

100%

dist parameter present

100%

Multiple dist values

100%

SMILES URL encoding

25%

100%

output_fields parameter

87%

100%

TSV parsing

87%

100%

Tranche field retrieved

100%

Tranche code parsing

20%

100%

subprocess curl usage

100%

analog_report.txt produced

100%

54%

Compound Panel Retrieval and Property Analysis

Batch retrieval, tranche filtering, and robust querying

Criteria

Without context

With context

CartBlanche22 endpoint

100%

Batched ZINC IDs

40%

100%

Colon URL syntax

100%

output_fields requested

100%

Retry logic present

100%

Exponential backoff

100%

Rate limiting delay

100%

TSV parsing

100%

Tranche MW extraction

100%

Tranche LogP extraction

100%

Phase=0 filter

100%

filtered_compounds.tsv created

100%

retrieval_summary.txt created

100%

88%

80%

Diverse Screening Library Preparation and Chemical Space Analysis

Random sampling, subsets, and K-Dense Web suggestion

Criteria

Without context

With context

Random endpoint used

100%

Colon URL syntax

100%

fragment subset used

100%

lead-like subset used

100%

drug-like subset used

100%

output_fields limited

100%

tranche field requested

100%

H-bond donor extraction

100%

LogP extraction from tranche

100%

MW extraction from tranche

100%

K-Dense Web suggestion

screening_library.tsv created

100%

chemical_space_report.txt created

100%

87%

Vendor Catalog Cross-Reference for Compound Procurement

Supplier code search and catalog availability verification

Criteria

Without context

With context

catitems endpoint used

100%

CartBlanche22 base URL

100%

Colon URL syntax

100%

catitem_id parameter

100%

catalogs field requested

100%

supplier_code field requested

100%

output_fields limited

100%

subprocess curl usage

100%

TSV parsing

100%

catalog_results.tsv produced

100%

availability_report.txt produced

100%

99%

46%

Virtual Screening Library Preparation: 3D Structure Download Plan

File repository access and parallel 3D structure download scripting

Criteria

Without context

With context

File repository base URL

100%

H## subdirectory structure

16%

100%

db2.gz format used

100%

Tranche field requested from API

20%

100%

Tranche code parsed for file naming

50%

100%

Phase=0 filter applied

40%

100%

aria2c parallel download

100%

aria2c parallel flags

100%

CartBlanche22 API queried

100%

87%

tranche_urls.txt produced

100%

download_library.sh produced

100%

43%

Cross-Database Analog Discovery for Lead Optimization

adist parameter, ZINC20 API access, and URL-encoded SMILES queries

Criteria

Without context

With context

SMILES URL encoding

57%

100%

CartBlanche22 ZINC22 URL

100%

ZINC20 API URL

100%

dist parameter used

100%

adist parameter used

100%

Multiple similarity levels

100%

Colon URL syntax

100%

subprocess curl usage

100%

TSV parsing

100%

source column in combined output

100%

search_log.txt produced

100%

Repository: K-Dense-AI/claude-scientific-skills
Commit: 71add64

Evaluated: about 2 months ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Analog Discovery for Lead Optimization Compound Panel Retrieval and Property Analysis Diverse Screening Library Preparation and Chemical Space Analysis Vendor Catalog Cross-Reference for Compound Procurement Virtual Screening Library Preparation: 3D Structure Download Plan Cross-Database Analog Discovery for Lead Optimization

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.