- Home
- Skills
- Research & Analysis
- Target scout: ChEMBL compounds, drugs and PubMed papers
Works with the AI Agents you already use
Target scout: ChEMBL compounds, drugs and PubMed papers
From a target name or UniProt ID: the ChEMBL target, most potent compounds (CSV), approved and clinical drugs, and recent PubMed papers.
$12
Target scout: ChEMBL compounds, drugs and PubMed papers
Example session with this skill installed
What's known about EGFR? Pull the potent compounds and the approved drugs.
- Read your context and instructions
- Compiled the target scout
- Generated the data export
Summary
Target: CHEMBL203 - Epidermal growth factor receptor (Homo sapiens, SINGLE PROTEIN). UniProt P00533, gene EGFR.
Found: 70 drugs at phase 1 or above (15 approved), the most potent compounds as a CSV, and 4 recent papers.
Approved drugs (15 of 70)
| ChEMBL ID | name | status | action |
|---|---|---|---|
| CHEMBL1173655 | AFATINIB | approved | INHIBITOR |
| CHEMBL4297774 | AMIVANTAMAB | approved | INHIBITOR |
| CHEMBL3545311 | BRIGATINIB | approved | INHIBITOR |
| CHEMBL1201577 | CETUXIMAB | approved | INHIBITOR |
| CHEMBL553 | ERLOTINIB | approved | INHIBITOR |
| CHEMBL939 | GEFITINIB | approved | INHIBITOR |
| CHEMBL554 | LAPATINIB | approved | INHIBITOR |
| CHEMBL3353410 | OSIMERTINIB | approved | INHIBITOR |
(7 more approved, plus 55 at phase 1-3, in the full report)
Potent compounds
Written to litscout_EGFR_compounds.csv with columns id, smiles, pchembl, standard_type, standard_value_nM, assay, year, document. Because it has a smiles and an id column, it feeds straight into a triage or ADMET run without editing.
Recent literature
The newest PubMed hits for "EGFR inhibitor", each with title, journal, year, PMID and link.
Next steps
- Triage the compound CSV to drop anything with PAINS or Brenk alerts before you model it.
- Note the approved list is split between small molecules (erlotinib, osimertinib) and antibodies (cetuximab, amivantamab); filter by molecule type if you only want chemical matter.
- Potency values come from different assays, so treat the ranking as a shortlist, not a leaderboard.
Files: litscout_EGFR.md, litscout_EGFR_compounds.csv
Real output from a live ChEMBL and PubMed run.
litscout_EGFR_compounds.csv
CSV · data export
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Ask your agent what is known about a target and get a sourced briefing back in under a minute.
What it does
- Resolve: maps a target name, gene symbol or UniProt accession to a ChEMBL entry, preferring the human SINGLE PROTEIN. This matters: a plain search for "EGFR" returns PPI complexes and the mouse protein above the human one.
- Potent compounds: pulls IC50/Ki activities in nM, keeps the best pChEMBL per compound, and writes a CSV with id, SMILES, pChEMBL, assay and year.
- Drug landscape: lists drugs at phase 1 and above with their mechanism and action, approved ones first.
- Literature: the newest PubMed papers for the target, with title, journal, year, PMID and link.
- Output: a Markdown report plus a compound CSV that feeds straight into triage or ADMET (the triage script is bundled).
Why it's trustworthy
Everything comes from the public ChEMBL REST API and NCBI E-utilities, fetched by a bundled, self-tested script (scripts/litscout.py --selftest); the model never recalls a compound or a paper from memory. If one source fails, the report says so rather than quietly dropping it.
Requirements
uv and internet access. Standard library only, no API keys, no paid data sources. Requests are rate-limited and identify themselves.
- Only covers what ChEMBL and PubMed hold: patents, preprints and unpublished data are not included.
- Picks one ChEMBL target, so drugs registered against a protein family rather than the single protein can be missed.
- Potency values come from different assays and labs; they are a shortlist, not a strict ranking.
- PubMed search is keyword-based ("<target> inhibitor"), not a systematic review.
- Needs internet access; if ChEMBL or NCBI is unreachable the report says so instead of guessing.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 1 month ago
Requires
Permissions
Allowed Hosts
Shell: runs the bundled scripts/litscout.py with uv. Network: queries the public ChEMBL REST API (ebi.ac.uk) and NCBI E-utilities; no keys and no other hosts. Read/Write: writes the report and compound CSV into the output folder you choose.