Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    ADMET risk screen: solubility, absorption, BBB, hERG

    1

    Rule-based ESOL solubility, Egan oral absorption, BBB, hERG and tox-risk flags, each with a concrete med-chem fix.

    $12

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    ADMET risk screen: solubility, absorption, BBB, hERG

    ADMET risk screen: solubility, absorption, BBB, hERG

    Example session with this skill installed

    Screen these 4 drugs for ADMET liabilities: terfenadine, imatinib, diazepam, caffeine.

    • Read your context and instructions
    • Compiled the admet risk screen
    • Generated the data export

    Script summary

    Screened 4 compounds (0 invalid): 1 clean, 3 flagged. Poor solubility 2 · poor oral absorption 1 · hERG liability 2 · Pfizer 3/75 risk 2.

    idlogS (ESOL)solubilityoral abs.BBBhERGflags
    terfenadine-6.68very lowpoorunlikelyhighpoor solubility; poor oral absorption; hERG liability; Pfizer 3/75 tox risk
    imatinib-5.81lowgoodunlikelyhighpoor solubility; hERG liability
    diazepam-3.97moderategoodlikelylowPfizer 3/75 tox risk
    caffeine-0.87highgoodunlikelylownone

    Liabilities and one fix each

    terfenadine, hERG: a basic piperidine plus cLogP > 3.7 is the classic hERG pharmacophore. Terfenadine was withdrawn over QT prolongation, so the rule reproduces a known outcome. Fix: lower amine basicity or add an acidic group (its successor fexofenadine is a zwitterion).

    terfenadine / imatinib, solubility: reduce cLogP or add a solubilising group. Imatinib's N-methylpiperazine already serves that role; salt forms help further.

    diazepam, Pfizer 3/75: cLogP > 3 with TPSA < 75 is linked to higher in-vivo toxicity odds. Expected for a CNS drug where BBB penetration is the goal, so adding polarity would trade away brain exposure.

    Only caffeine is clean on every filter.

    These are heuristic prioritization filters (Delaney ESOL, Egan, BBB rules, basic-amine hERG flag, Pfizer 3/75, GSK 4/400), not measured or ML-predicted ADMET. Use them to decide what to send to assays.

    Full table: admet.csv

    admet.csv

    CSV · data export

    Generated

    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

    Screens SMILES or a compound CSV for developability liabilities using published, citable rules: Delaney ESOL, Egan egg, BBB rules, a basic-amine hERG flag, Pfizer 3/75 and GSK 4/400. Every flag comes with a suggested medicinal-chemistry fix.

    What it does

    • Solubility: Delaney ESOL logS, banded high / moderate / low / very low.
    • Oral absorption: Egan egg (TPSA <= 131.6 and cLogP <= 5.88).
    • BBB: TPSA < 90, MW < 450, HBD <= 3, 1 <= cLogP <= 4.
    • hERG: basic aliphatic amine plus cLogP > 3.7 (high) or > 2.5 (medium).
    • Tox / developability: Pfizer 3/75 and GSK 4/400.
    • Output: a full CSV plus a summary that leads with the flagged compounds.

    Why it's trustworthy

    Every descriptor comes from a bundled, self-tested RDKit script (scripts/admet.py --selftest); the model never estimates a property. The agent reports the flags and, for each one, a concrete structural fix.

    These are heuristic filters for prioritization, and the skill says so in its own output. They are not measured or ML-predicted ADMET.

    Requirements

    uv. RDKit installs automatically on first run. Accepts CSV, .smi/.txt or a single SMILES string. The compound-triage script is bundled, so triage and ADMET can be joined on id.

    • Heuristic prioritization filters (Delaney ESOL, Egan, BBB rules, basic-amine hERG flag, Pfizer 3/75, GSK 4/400), not measured or ML-predicted ADMET.
    • No metabolism, CYP, transporter or PK prediction.
    • Rules are built for small molecules; peptides, macrocycles and biologics fall outside their validated range.
    • Needs uv and one-time internet access to install RDKit.

    How to use

    1. 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. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 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

    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access

    Shell: runs the bundled scripts/admet.py with uv. Network: only uv downloading RDKit on first run. Read/Write: reads your input CSV/.smi and writes the ADMET CSV you name with -o.

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    The skill accepts single SMILES strings for quick analysis or batch processing via .smi and .csv files, making it suitable for both individual molecule checks and large library screenings.

    The skill runs RDKit to compute the descriptors (MW, cLogP, TPSA, HBD/HBA, rotatable bonds) and applies the published rules to those numbers. The model never estimates a property from the SMILES string, so the same input gives the same result every run.

    You receive the core Python screening script, the logic for all rule-based filters (e.g., Pfizer 3/75, GSK 4/400), and the necessary environment configurations for the 'uv' package manager.

    This skill requires the 'uv' package manager to handle dependencies. It is designed to be compatible with any AI agent or platform capable of executing Python scripts within a containerized or local environment.

    Yes, for every liability flagged (such as low solubility or hERG risk), the skill provides specific medicinal chemistry suggestions on how to modify the structure to improve its ADMET profile.

    These are rule-based heuristic filters used for early-stage prioritization; they provide high-speed screening but should be followed by experimental assays or high-fidelity ML simulations for final validation.

    New in Research & Analysis