{
  "schemaVersion": "1.0",
  "item": {
    "slug": "admet-prediction",
    "name": "Admet Prediction",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/huifer/admet-prediction",
    "canonicalUrl": "https://clawhub.ai/huifer/admet-prediction",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/admet-prediction",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=admet-prediction",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "SKILL.md",
      "reference/admet-properties.md",
      "reference/toxicity-alerts.md",
      "scripts/admet_predict.py"
    ],
    "primaryDoc": "SKILL.md",
    "quickSetup": [
      "Download the package from Yavira.",
      "Extract the archive and review SKILL.md first.",
      "Import or place the package into your OpenClaw setup."
    ],
    "agentAssist": {
      "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.",
      "steps": [
        "Download the package from Yavira.",
        "Extract it into a folder your agent can access.",
        "Paste one of the prompts below and point your agent at the extracted folder."
      ],
      "prompts": [
        {
          "label": "New install",
          "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete."
        },
        {
          "label": "Upgrade existing",
          "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run."
        }
      ]
    },
    "sourceHealth": {
      "source": "tencent",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-23T16:43:11.935Z",
      "expiresAt": "2026-04-30T16:43:11.935Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard",
        "contentDisposition": "attachment; filename=\"4claw-imageboard-1.0.1.zip\"",
        "redirectLocation": null,
        "bodySnippet": null
      },
      "scope": "source",
      "summary": "Source download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this source.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/admet-prediction"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    },
    "downloadPageUrl": "https://openagent3.xyz/downloads/admet-prediction",
    "agentPageUrl": "https://openagent3.xyz/skills/admet-prediction/agent",
    "manifestUrl": "https://openagent3.xyz/skills/admet-prediction/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/admet-prediction/agent.md"
  },
  "agentAssist": {
    "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.",
    "steps": [
      "Download the package from Yavira.",
      "Extract it into a folder your agent can access.",
      "Paste one of the prompts below and point your agent at the extracted folder."
    ],
    "prompts": [
      {
        "label": "New install",
        "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete."
      },
      {
        "label": "Upgrade existing",
        "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run."
      }
    ]
  },
  "documentation": {
    "source": "clawhub",
    "primaryDoc": "SKILL.md",
    "sections": [
      {
        "title": "ADMET Prediction Skill",
        "body": "Predict ADMET properties to prioritize compounds for development."
      },
      {
        "title": "Quick Start",
        "body": "/admet \"CC1=CC=C(C=C1)CNC\" --full\n/pk-prediction --library compounds.sdf --threshold 0.7\n/toxicity-screen CHEMBL210 --include hERG,DILI,Ames"
      },
      {
        "title": "What's Included",
        "body": "PropertyPredictionModelAbsorptionCaco-2, HIA, PgpML/QSARDistributionVDss, PPB, BBBML/QSARMetabolismCYP inhibition, clearanceML/QSARExcretionClearance, half-lifeML/QSARToxicityhERG, DILI, Ames, mutagenicityML/QSAR"
      },
      {
        "title": "Output Structure",
        "body": "# ADMET Profile: CHEMBL210 (Osimertinib)\n\n## Summary\n| Property | Value | Status |\n|----------|-------|--------|\n| Drug-likeness | Pass | ✓ |\n| Lipinski Ro5 | 0 violations | ✓ |\n| VEBER | Pass | ✓ |\n| PAINS | 0 alerts | ✓ |\n| Brenk | 0 alerts | ✓ |\n\n## Absorption\n| Property | Prediction | Confidence |\n|----------|------------|-------------|\n| HIA | 98% | High |\n| Caco-2 | 15.2 × 10⁻⁶ cm/s | High |\n| Pgp substrate | Yes | Medium |\n| F30% | 65% | Medium |\n\n## Distribution\n| Property | Prediction | Confidence |\n|----------|------------|-------------|\n| VDss | 5.2 L/kg | Medium |\n| PPB | 95% | High |\n| BBB | Yes | High |\n| CNS MPO | 5.5 | Good |\n\n## Metabolism\n| Property | Prediction | Confidence |\n|----------|------------|-------------|\n| CYP3A4 substrate | Yes | High |\n| CYP3A4 inhibitor | Yes | Medium |\n| CYP2D6 inhibitor | No | High |\n| CYP2C9 inhibitor | No | Medium |\n| Clearance | 8.5 mL/min/kg | Low |\n\n## Excretion\n| Property | Prediction | Confidence |\n|----------|------------|-------------|\n| Renal clearance | 10% | Medium |\n| Half-life | 48 hours | High |\n\n## Toxicity\n| Property | Prediction | Confidence |\n|----------|------------|-------------|\n| hERG inhibition | No | High |\n| DILI | Concern | Medium |\n| Ames mutagenicity | Negative | High |\n| Carcinogenicity | Negative | Medium |\n| Respiratory toxicity | No | Low |\n\n## Recommendations\n**Strengths**:\n- Good oral bioavailability (65%)\n- Brain penetration (BBB permeable)\n- Low hERG risk\n\n**Concerns**:\n- DILI concern - monitor in preclinical studies\n- CYP3A4 inhibition - potential DDIs\n\n**Overall**: Good ADMET profile. Progress to in vivo PK."
      },
      {
        "title": "Drug-Likeness",
        "body": "RulePass CriteriaLipinski Ro5≤ 1 violationVeberRotB ≤ 10, PSA ≤ 140 ŲEganLogP ≤ 5, PSA ≤ 131 ŲMDDRMW ≤ 600, LogP ≤ 5"
      },
      {
        "title": "Absorption",
        "body": "PropertyGoodModeratePoorHIA>80%40-80%<40%Caco-2>101-10<1F30%>70%30-70%<30%"
      },
      {
        "title": "Distribution",
        "body": "PropertyGoodModeratePoorVDss0.3-5 L/kg<0.3 or >5ExtremePPB<90%90-95%>95%BBBLogBB > 0.3-0.3 to 0.3< -0.3"
      },
      {
        "title": "Toxicity Alerts",
        "body": "AlertActionhERG inhibitionCardiotoxicity riskDILI positiveHepatotoxicity riskAmes positiveMutagenicity riskPAINSAssay interferenceStructural alertsInvestigate further"
      },
      {
        "title": "Running Scripts",
        "body": "# Full ADMET profile\npython scripts/admet_predict.py --smiles \"CC1=CC=C...\" --full\n\n# Batch prediction\npython scripts/admet_predict.py --library compounds.sdf --output results.csv\n\n# Specific properties\npython scripts/admet_predict.py --smiles \"...\" --properties hERG,DILI,CYP\n\n# Filter by criteria\npython scripts/admet_filter.py --library compounds.sdf --rules lipinski,veber"
      },
      {
        "title": "Requirements",
        "body": "pip install rdkit\n\n# Optional for advanced models\npip install deepchem admet-x"
      },
      {
        "title": "Reference",
        "body": "reference/admet-properties.md - Detailed property reference\nreference/toxicity-alerts.md - Toxicity alerts reference\nreference/pk-models.md - PK prediction models"
      },
      {
        "title": "Best Practices",
        "body": "Use multiple models: Consensus predictions more reliable\nCheck confidence: Low confidence = experimental verification needed\nConsider chemistry: Novel structures less reliable\nIterative design: Use predictions to guide synthesis\nValidate early: Confirm key predictions experimentally"
      },
      {
        "title": "Common Pitfalls",
        "body": "PitfallSolutionOver-reliance on predictionsExperimental validation requiredIgnoring confidenceCheck model applicability domainSingle model onlyUse consensus of multiple modelsIgnoring chemistryNovel scaffolds = uncertain predictionsLate-stage testingEarly ADMET screening saves time"
      },
      {
        "title": "Limitations",
        "body": "Models are approximate: Errors common\nNovel chemistry: Less reliable for new scaffolds\nIn vitro-in vivo gap: Predictions don't always translate\nSpecies differences: Human predictions based on animal data\nComplex mechanisms: Some toxicity not predicted"
      }
    ],
    "body": "ADMET Prediction Skill\n\nPredict ADMET properties to prioritize compounds for development.\n\nQuick Start\n/admet \"CC1=CC=C(C=C1)CNC\" --full\n/pk-prediction --library compounds.sdf --threshold 0.7\n/toxicity-screen CHEMBL210 --include hERG,DILI,Ames\n\nWhat's Included\nProperty\tPrediction\tModel\nAbsorption\tCaco-2, HIA, Pgp\tML/QSAR\nDistribution\tVDss, PPB, BBB\tML/QSAR\nMetabolism\tCYP inhibition, clearance\tML/QSAR\nExcretion\tClearance, half-life\tML/QSAR\nToxicity\thERG, DILI, Ames, mutagenicity\tML/QSAR\nOutput Structure\n# ADMET Profile: CHEMBL210 (Osimertinib)\n\n## Summary\n| Property | Value | Status |\n|----------|-------|--------|\n| Drug-likeness | Pass | ✓ |\n| Lipinski Ro5 | 0 violations | ✓ |\n| VEBER | Pass | ✓ |\n| PAINS | 0 alerts | ✓ |\n| Brenk | 0 alerts | ✓ |\n\n## Absorption\n| Property | Prediction | Confidence |\n|----------|------------|-------------|\n| HIA | 98% | High |\n| Caco-2 | 15.2 × 10⁻⁶ cm/s | High |\n| Pgp substrate | Yes | Medium |\n| F30% | 65% | Medium |\n\n## Distribution\n| Property | Prediction | Confidence |\n|----------|------------|-------------|\n| VDss | 5.2 L/kg | Medium |\n| PPB | 95% | High |\n| BBB | Yes | High |\n| CNS MPO | 5.5 | Good |\n\n## Metabolism\n| Property | Prediction | Confidence |\n|----------|------------|-------------|\n| CYP3A4 substrate | Yes | High |\n| CYP3A4 inhibitor | Yes | Medium |\n| CYP2D6 inhibitor | No | High |\n| CYP2C9 inhibitor | No | Medium |\n| Clearance | 8.5 mL/min/kg | Low |\n\n## Excretion\n| Property | Prediction | Confidence |\n|----------|------------|-------------|\n| Renal clearance | 10% | Medium |\n| Half-life | 48 hours | High |\n\n## Toxicity\n| Property | Prediction | Confidence |\n|----------|------------|-------------|\n| hERG inhibition | No | High |\n| DILI | Concern | Medium |\n| Ames mutagenicity | Negative | High |\n| Carcinogenicity | Negative | Medium |\n| Respiratory toxicity | No | Low |\n\n## Recommendations\n**Strengths**:\n- Good oral bioavailability (65%)\n- Brain penetration (BBB permeable)\n- Low hERG risk\n\n**Concerns**:\n- DILI concern - monitor in preclinical studies\n- CYP3A4 inhibition - potential DDIs\n\n**Overall**: Good ADMET profile. Progress to in vivo PK.\n\nProperty Ranges\nDrug-Likeness\nRule\tPass Criteria\nLipinski Ro5\t≤ 1 violation\nVeber\tRotB ≤ 10, PSA ≤ 140 Ų\nEgan\tLogP ≤ 5, PSA ≤ 131 Ų\nMDDR\tMW ≤ 600, LogP ≤ 5\nAbsorption\nProperty\tGood\tModerate\tPoor\nHIA\t>80%\t40-80%\t<40%\nCaco-2\t>10\t1-10\t<1\nF30%\t>70%\t30-70%\t<30%\nDistribution\nProperty\tGood\tModerate\tPoor\nVDss\t0.3-5 L/kg\t<0.3 or >5\tExtreme\nPPB\t<90%\t90-95%\t>95%\nBBB\tLogBB > 0.3\t-0.3 to 0.3\t< -0.3\nToxicity Alerts\nAlert\tAction\nhERG inhibition\tCardiotoxicity risk\nDILI positive\tHepatotoxicity risk\nAmes positive\tMutagenicity risk\nPAINS\tAssay interference\nStructural alerts\tInvestigate further\nRunning Scripts\n# Full ADMET profile\npython scripts/admet_predict.py --smiles \"CC1=CC=C...\" --full\n\n# Batch prediction\npython scripts/admet_predict.py --library compounds.sdf --output results.csv\n\n# Specific properties\npython scripts/admet_predict.py --smiles \"...\" --properties hERG,DILI,CYP\n\n# Filter by criteria\npython scripts/admet_filter.py --library compounds.sdf --rules lipinski,veber\n\nRequirements\npip install rdkit\n\n# Optional for advanced models\npip install deepchem admet-x\n\nReference\nreference/admet-properties.md - Detailed property reference\nreference/toxicity-alerts.md - Toxicity alerts reference\nreference/pk-models.md - PK prediction models\nBest Practices\nUse multiple models: Consensus predictions more reliable\nCheck confidence: Low confidence = experimental verification needed\nConsider chemistry: Novel structures less reliable\nIterative design: Use predictions to guide synthesis\nValidate early: Confirm key predictions experimentally\nCommon Pitfalls\nPitfall\tSolution\nOver-reliance on predictions\tExperimental validation required\nIgnoring confidence\tCheck model applicability domain\nSingle model only\tUse consensus of multiple models\nIgnoring chemistry\tNovel scaffolds = uncertain predictions\nLate-stage testing\tEarly ADMET screening saves time\nLimitations\nModels are approximate: Errors common\nNovel chemistry: Less reliable for new scaffolds\nIn vitro-in vivo gap: Predictions don't always translate\nSpecies differences: Human predictions based on animal data\nComplex mechanisms: Some toxicity not predicted"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/huifer/admet-prediction",
    "publisherUrl": "https://clawhub.ai/huifer/admet-prediction",
    "owner": "huifer",
    "version": "0.1.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/admet-prediction",
    "downloadUrl": "https://openagent3.xyz/downloads/admet-prediction",
    "agentUrl": "https://openagent3.xyz/skills/admet-prediction/agent",
    "manifestUrl": "https://openagent3.xyz/skills/admet-prediction/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/admet-prediction/agent.md"
  }
}