chembl-database
Query the ChEMBL database for bioactive molecules, drug targets, bioactivity data, approved drugs, and chemical structures
Install / Use
npx skills add google-deepmind/science-skills --skill chembl_databaseInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
CommunicationSupported Platforms
Our assessment of chembl-database
chembl-database scores 93/100 on our quality scale, 118th of 394 Communication skills we index (top 30%).
Its SKILL.md is 11 KB long, well organised into 19 sections with 5 code examples: a thorough specification that gives an agent plenty to work with.
With 3,137 GitHub stars, it is one of the more widely adopted skills in the catalogue.
Maintenance, license and trust
- The repository was last updated about 2 months ago, so chembl-database is actively maintained.
- Our last check on 2026-10-26 found the source still online.
- It is released under the Apache-2.0 license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-09-29. Automated pattern scan on 2026-09-29. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
chembl-database compared with similar skills
All 4 of these similar skills score higher than chembl-database; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| chembl-database (this skill)by google-deepmind | 93 | 3.1k | 2mo ago | SKILL.md |
| x402by browser-use | 100 | 121.0k | 3d ago | SKILL.md |
| algorithmic-artby anthropics | 100 | 177.9k | 55d ago | SKILL.md |
| pptxby anthropics | 100 | 177.9k | 55d ago | SKILL.md |
| designby nextlevelbuilder | 100 | 149.0k | 6d ago | SKILL.md |
Frequently asked questions
- How do I install chembl-database?
- Run
npx skills add google-deepmind/science-skills --skill chembl-database. The install tabs above show the steps for each supported agent. - Which AI agents does chembl-database work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is chembl-database safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is Apache-2.0-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is chembl-database still maintained?
- The repository was last updated about 2 months ago, so chembl-database is actively maintained.
Skill content
View source on GitHubname: chembl-database description: > Query the ChEMBL database for bioactive molecules, drug targets, bioactivity data, approved drugs, and chemical structures. Use when the user asks about compounds, targets, IC50/Ki values, drug mechanisms, or structure searches.
ChEMBL Database Query
Prerequisites
uv: Read theuvskill and follow its Setup instructions to ensureuvis installed and on PATH.- User Notification: If .licenses/chembl_database_LICENSE.txt does not already exist in the workspace root directory then (1) prominently notify the user to check the terms at https://chembl.gitbook.io/chembl-interface-documentation/about, then (2) create the file recording the notification text and timestamp.
Core Rules
-
[!IMPORTANT] Use the Utility Scripts: You MUST ALWAYS use the provided utility script
scripts/chembl_api.pyfor all ChEMBL API interactions, including checking status. NEVER usecurlor custom Python requests to query the ChEMBL API directly. This ensures rate limit is enfoced and also retries on network errors. -
Output to File (Required): The
--outputflag is required for every subcommand. All JSON results are written to the specified file. After running the command, read the output file with jq or your own code to extract the data. List results are typically wrapped in a JSON array keyed by the endpoint name (e.g.,molecules,activities). -
Notification: If this skill is used, ensure this is mentioned in the output.
Utility Script
All ChEMBL API queries use one script with subcommands:
uv run scripts/chembl_api.py <subcommand> --output <file> [options]
1. Check API Status
uv run scripts/chembl_api.py status --output /tmp/status.json
2. Molecule Queries
Fetch by ChEMBL ID: bash uv run scripts/chembl_api.py molecule --id CHEMBL25 --output /tmp/mol.json
Search by name: bash uv run scripts/chembl_api.py molecule --search "aspirin" --limit 3 --output /tmp/mol_search.json
Batch fetch: bash uv run scripts/chembl_api.py molecule --ids "CHEMBL25;CHEMBL1642" --limit 10 --output /tmp/mol_batch.json
Filter by properties: bash uv run scripts/chembl_api.py molecule --filter molecule_properties__mw_freebase__lte=500 --limit 5 --output /tmp/mol_filter.json
Filter by range: bash uv run scripts/chembl_api.py molecule --filter molecule_properties__mw_freebase__range=150,200 --limit 5 --output /tmp/mol_range.json
Download SDF structure file: bash uv run scripts/chembl_api.py molecule --id CHEMBL25 --dl_format sdf --output /tmp/aspirin.sdf
Tip: SDF/MOL files can be passed directly to tools like PyMOL or RDKit for 3D visualization and analysis.
3. Target Queries
Search for targets: bash uv run scripts/chembl_api.py target --search "EGFR" --limit 5 --output /tmp/targets.json
Fetch by ID: bash uv run scripts/chembl_api.py target --id CHEMBL203 --output /tmp/egfr.json
4. Bioactivity Data
Fetch activity by ID: bash uv run scripts/chembl_api.py activity --id 31863 --output /tmp/act.json
Search activities: bash uv run scripts/chembl_api.py activity --search "EGFR" --limit 5 --output /tmp/act_search.json
Filter activities for a target: bash uv run scripts/chembl_api.py activity --filter target_chembl_id=CHEMBL203 standard_type=IC50 --limit 10 --output /tmp/egfr_ic50.json
Normalize bioactivity units to nM: bash uv run scripts/chembl_api.py activity --filter target_chembl_id=CHEMBL203 standard_type=IC50 --limit 5 --normalize --output /tmp/egfr_normalized.json
Important: Bioactivity values come in various units (nM, µM, pM). Use
--normalizeto convert all values to nM for consistent comparison. Each record will includenormalized_value_nMandnormalization_note.
5. Drug Information
Fetch drug details: bash uv run scripts/chembl_api.py drug --id CHEMBL25 --output /tmp/drug.json
Drug indications: bash uv run scripts/chembl_api.py drug_indication --filter molecule_chembl_id=CHEMBL25 --limit 10 --output /tmp/indications.json
Filter indications by phase: bash uv run scripts/chembl_api.py drug_indication --filter molecule_chembl_id=CHEMBL25 max_phase_for_ind=4.0 --limit 10 --output /tmp/approved_indications.json
Drug warnings: bash uv run scripts/chembl_api.py drug_warning --limit 5 --output /tmp/warnings.json
Mechanisms of action: bash uv run scripts/chembl_api.py mechanism --filter molecule_chembl_id=CHEMBL25 --limit 5 --output /tmp/mech.json
6. Structure-Based Searches
Note: Both similarity and substructure searches are performed server-side on ChEMBL's pre-indexed database. They do not require a local RDKit installation.
Similarity search (SMILES + threshold): bash uv run scripts/chembl_api.py similarity --smiles "CC(=O)Oc1ccccc1C(=O)O" --similarity 85 --limit 5 --output /tmp/similar.json
Substructure search (SMILES): bash uv run scripts/chembl_api.py substructure --smiles "c1ccccc1" --limit 5 --output /tmp/substruct.json
7. Compound Image
Download a 2D structure image (SVG by default, scalable for publication):
uv run scripts/chembl_api.py image --id CHEMBL25 --output /tmp/chembl25.svg
Options:
--dimensions: Image size in pixels (max 500, default 500).--engine: Rendering engine (default: rdkit).--img_format: Output format —svg(default, vector) orpng(raster).
8. Cross-Referencing with Other Databases
ChEMBL integrates with UniProt, Ensembl, PubChem, and other databases. Common cross-referencing patterns:
Find a ChEMBL target from a UniProt accession: bash uv run scripts/chembl_api.py target --filter target_components__accession=P00533 --limit 5 --output /tmp/uniprot_target.json
Resolve any ChEMBL ID to its entity type: bash uv run scripts/chembl_api.py chembl_id_lookup --id CHEMBL203 --output /tmp/lookup.json
Look up cross-reference sources: bash uv run scripts/chembl_api.py xref_source --limit 10 --output /tmp/xrefs.json
Tip: Use the
target_componentendpoint to find UniProt accessions, gene names, and protein sequences for any ChEMBL target.
9. Pagination
All list endpoints support --limit and --offset for pagination:
# First page: 2 results starting at offset 0
uv run scripts/chembl_api.py molecule --limit 2 --offset 0 --output /tmp/page1.json
# Second page: next 2 results starting at offset 2
uv run scripts/chembl_api.py molecule --limit 2 --offset 2 --output /tmp/page2.json
The response includes page_meta with total_count, limit, offset, next,
and previous links. Use successive --offset values to page through large
result sets.
10. Other Endpoints
All remaining endpoints follow the same pattern:
uv run scripts/chembl_api.py <subcommand> --output <file> [--id ID | --ids ID1;ID2 | --search QUERY] [--limit N] [--offset N] [--filter KEY=VAL ...]
Key subcommands at a glance:
molecule(searchable: true): Molecules/compounds — the primary entry pointtarget(searchable: true): Drug targets (proteins, organisms, etc.)activity(searchable: true): Bioactivity data (IC50, Ki, EC50, etc.)drug(searchable: false): Approved drugsmechanism(searchable: false): Mechanisms of actionassay(searchable: true): Assay descriptionssimilarity(searchable: false): Similarity search (special)substructure(searchable: false): Substructure search (special)image(searchable: false): Compound image download (special)
Full subcommand list:
activity_supp(searchable: false): Supplementary activity dataassay_class(searchable: false): Assay classificationsatc_class(searchable: false): ATC drug classificationsbinding_site(searchable: false): Binding site informationbiotherapeutic(searchable: false): Biotherapeutic moleculescell_line(searchable: false): Cell line detailschembl_id_lookup(searchable: true): ChEMBL ID resolutionchembl_release(searchable: false): Database release infocompound_record(searchable: false): Compound recordscompound_structural_alert(searchable: false): Structural alertsdocument(searchable: true): Literature documentsdocument_similarity(searchable: false): Document similaritydrug_indication(searchable: false): Drug indicationsdrug_warning(searchable: false): Drug safety warningsgo_slim(searchable: false): GO slim termsmetabolism(searchable: false): Metabolism datamolecule_form(searchable: false): Molecule forms (salts/parents)organism(searchable: false): Organismsprotein_classification(searchable: true): Protein classificationssource(searchable: false): Data sourcestarget_component(searchable: false): Target protein componentstarget_relation(searchable: false): Target relationshipstissue(searchable: false): Tissue typesxref_source(searchable: false): Cross-reference sourcesstatus(searchable: false): API status check (special)
Common Options
--output FILE: Required. Output file path for JSON results.--id ID: Fetch a single record by ID.--ids ID1;ID2;...: Batch fetch multiple records.--search QUERY: Free-text search (only for searchable endpoints, marked ✓).--limit N: Max results to return (default: 5).--offset N: Pagination offset.--filter KEY=VAL: Filter parameters (can specify multiple).--normalize: (activity only) Normalize values to nM.--dl_format sdf|mol: (molecule only) Download structure file.
Reference
- API Endpoints Reference: See references/api_endpoints.md for the full list of endpoints and filter operators.
Workflow
- Use
status --output /tmp/status.jsonto verify the API is available. - Search for targets, molecules, or drugs using the relevant subcommand.
- Read the output JSON file to extract IDs and data.
- Use IDs from search results to fetch detailed records.
- Query
activitywith filters to get bioactivity data for targets/molecules. Use--normalizewhen comparing values across studies. - Use
similarityorsubstructurefor server-side structure-based queries. - Download compound images with
imageor structure files withmolecule --dl_format sdf. - Use
target --filter target_components__accession=<UniProt>to cross- reference with UniProt.
Related Skills
x402
121.0kSet up Browser Use Cloud payments with x402 — pay per request from a crypto wallet (USDC on Base mainnet), no signup or API key.
algorithmic-art
177.9kCreating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems.
pptx
177.9kUse this skill any time a .pptx or .potx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx or .potx file (even if the extracted content will be used elsewhere, like in an em…
design
149.0kComprehensive design skill: brand identity, design tokens, UI styling, logo generation (55 styles, Gemini, Atlas Cloud, or MuAPI AI), corporate identity program (50 deliverables, CIP mockups), HTML presentations (Chart.js), banner design (22 styles, social/ads/web/print), icon design (15 styles, SVG…
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
