chembl-molecule-search

SkillSearch

Use this skill to search ChEMBL molecules by keyword and return a compact compound summary. Do not use it for local chemistry calculations or unsupported bulk export.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the chembl-molecule-search skill

What this skill tells your AI

The instructions your AI receives, as published by ma-compbio-lab/skillfoundry in skills/drug-discovery-and-cheminformatics/chembl-molecule-search/SKILL.md and read by ahel’s review.

Purpose

Search ChEMBL by free-text query and return compact compound summaries suitable for quick cheminformatics triage.

When to use

  • You need a ChEMBL identifier and basic compound summary for a known drug or molecule name.
  • You want a lightweight compound lookup before deeper cheminformatics or screening workflows.

When not to use

  • You need local descriptor calculation or docking.
  • You need bulk export of a large ChEMBL slice.
  • You need offline execution.

Inputs

  • Query string
  • Optional result limit and output path

Outputs

  • JSON payload with query metadata and compact molecule summaries

Requirements

  • Python 3.10+
  • Network access to www.ebi.ac.uk

Procedure

  1. Run python3 skills/drug-discovery-and-cheminformatics/chembl-molecule-search/scripts/search_chembl_molecules.py --query imatinib --limit 1 --out skills/drug-discovery-and-cheminformatics/chembl-molecule-search/assets/chembl_imatinib.json.
  2. Inspect molecules for chembl_id, preferred_name, molecule_type, max_phase, canonical_smiles, and selected properties.
  3. Use the returned ChEMBL IDs in downstream compound-knowledge workflows when needed.

Validation

  • Command exits successfully.
  • molecules contains at least one ChEMBL result for a known query.
  • The first result has a non-empty chembl_id and molecule summary fields.

Failure modes and fixes

  • No hits: refine the query or use a more canonical drug name.
  • Too many hits: lower --limit and inspect the top results manually.
  • HTTP errors: retry later and keep the query simple for smoke verification.

Safety and limits

  • Knowledge lookup only.
  • This skill does not perform synthesis planning, structure optimization, or wet-lab decision making.

Example

  • python3 skills/drug-discovery-and-cheminformatics/chembl-molecule-search/scripts/search_chembl_molecules.py --query imatinib --limit 1

Provenance

Related skills

  • clinicaltrials-study-search

Signals

GitHub stars
39
Forks
5
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
chembl-molecule-search-ma-compbio-lab
Source
github.com/ma-compbio-lab/skillfoundry