以下为本文档的中文说明tooluniverse-chemical-safety化学安全工具宇宙是由哈佛大学医学院 MIMS生物医学信息学实验室开发的一个专注于化学物质安全数据查询和危险评估的专业技能。该技能整合了多个权威的化学安全数据库为化学、生物、制药和环境等领域的科研人员和工业从业者提供便捷、全面、可靠的化学安全信息咨询。使用场景包括在实验设计阶段查询将要使用化学物质的物理化学性质参数、毒性等级和安全操作规程以制定实验安全方案、评估工业生产过程或实验室操作中涉及的化学品的火灾、爆炸和健康危害风险、在化学品泄漏、火灾等安全事故应急响应中快速查询和获取关键的安全处置信息和急救指导、为新化学品的采购、入库存储和日常使用提供科学的安全存储建议。核心特点包括整合查询多个国际权威化学安全数据源的分类信息包括联合国 GHS全球化学品统一分类和标签制度的危险分类和象形图、美国 NFPA 704美国消防协会的菱形标识健康/燃烧/反应/特殊危害、以及通过 CAS 号美国化学文摘社登记号精确检索化学品的唯一标识信息提供化学品全面的危险分类细节和危险性说明H 代码包括易燃性闪点、自燃温度、急性毒性LD50/LC50 口服/吸入/经皮、皮肤和眼腐蚀性/刺激性、严重危害水生环境等完整的安全数据表SDS/MSDS核心内容自动摘要提取包括第 4 部分急救措施眼睛接触、皮肤接触、吸入、食入后的处理、第 5 部分消防措施适用灭火介质、防护装备、有害燃烧产物、第 6 部分泄漏应急处理个人预防措施、环境保护措施、收容和清除方法支持化学结构式SMILES 字符串和系统命名/IUPAC 命名/IUPAC 名称的灵活搜索提供化学物质间的兼容性矩阵查询为安全存储隔离储存条件和正确混合使用提供关键决策支持。该技能是实验室安全管理和工业过程安全管理的重要参考工具。Chemical Safety Toxicology AssessmentToxicity assessment: identify the chemical, check known hazards (GHS, IARC), then look for ADMET predictions. Dose makes the poison — always consider exposure level, as a compound that is toxic at high doses may be safe at relevant exposures. Distinguish between acute toxicity (LD50, GHS category) and chronic hazards (carcinogenicity, endocrine disruption) — they require different risk management approaches. Computational predictions (ADMETAI) are T3 evidence and must be anchored by experimental data from PubChemTox or FDA labels wherever available. When evidence conflicts between prediction and experiment, always defer to the experimental finding.LOOK UP DON’T GUESS: never assume GHS categories, IARC classification, or CTD disease links — always call PubChemTox and CTD tools to retrieve current classifications before reporting.Comprehensive chemical safety analysis integrating predictive AI models, curated toxicogenomics databases, regulatory safety data, and chemical-biological interaction networks.When to Use This SkillTriggers:“Is this chemical toxic?” / “Assess the safety profile of [drug/chemical]”“What are the ADMET properties of [SMILES]?”“What genes does [chemical] interact with?” / “What diseases are linked to [chemical] exposure?”“Drug safety assessment” / “Environmental health risk” / “Chemical hazard profiling”Use Cases:Predictive Toxicology: AI-predicted endpoints (AMES, DILI, LD50, carcinogenicity, hERG) via SMILESADMET Profiling: Absorption, distribution, metabolism, excretion, toxicityToxicogenomics: Chemical-gene-disease mapping from CTDRegulatory Safety: FDA label warnings, contraindications, adverse reactionsDrug Safety: DrugBank safety FDA labels combinedChemical-Protein Interactions: STITCH-based interaction networksEnvironmental Toxicology: Chemical-disease associations for contaminantsCOMPUTE, DON’T DESCRIBEWhen analysis requires computation (statistics, data processing, scoring, enrichment), write and run Python code via Bash. Don’t describe what you would do — execute it and report actual results. Use ToolUniverse tools to retrieve data, then Python (pandas, scipy, statsmodels, matplotlib) to analyze it.KEY PRINCIPLESReport-first approach- Create report file FIRST, then populate progressivelyTool parameter verification- Verify params viaget_tool_infobefore calling unfamiliar toolsEvidence grading- Grade all safety claims by evidence strength (T1-T4)Citation requirements- Every toxicity finding must have inline source attributionMandatory completeness- All sections must exist with data or explicit “No data” notesDisambiguation first- Resolve compound identity (name - SMILES, CID, ChEMBL ID) before analysisNegative results documented- “No toxicity signals found” is data; empty sections are failuresConservative risk assessment- When evidence is ambiguous, flag as “requires further investigation”English-first queries- Always use English chemical/drug names in tool callsEvidence Grading System (MANDATORY)TierSymbolCriteriaExamplesT1[T1]Direct human evidence, regulatory findingFDA boxed warning, clinical trial toxicityT2[T2]Animal studies, validated in vitroNonclinical toxicology, AMES positive, animal LD50T3[T3]Computational prediction, association dataADMET-AI prediction, CTD associationT4[T4]Database annotation, text-minedLiterature mention, unvalidated database entryEvidence grades MUST appear in: Executive Summary, Toxicity Predictions, Regulatory Safety, Chemical-Gene Interactions, Risk Assessment.Core Strategy: 8 Research PhasesChemical/Drug Query | -- PHASE 0: Compound Disambiguation (ALWAYS FIRST) | Resolve name - SMILES, PubChem CID, ChEMBL ID, formula, weight | -- PHASE 1: Predictive Toxicology (ADMET-AI) | AMES, DILI, ClinTox, carcinogenicity, LD50, hERG, skin reaction | Stress response pathways, nuclear receptor activity | -- PHASE 2: ADMET Properties | BBB penetrance, bioavailability, clearance, CYP interactions, physicochemical | -- PHASE 3: Toxicogenomics (CTD) | Chemical-gene interactions, chemical-disease associations | -- PHASE 4: Regulatory Safety (FDA Labels) | Boxed warnings, contraindications, adverse reactions, nonclinical tox | -- PHASE 5: Drug Safety Profile (DrugBank) | Toxicity data, contraindications, drug interactions | -- PHASE 6: Chemical-Protein Interactions (STITCH) | Direct binding, off-target effects, interaction confidence | -- PHASE 7: Structural Alerts (ChEMBL) | PAINS, Brenk, Glaxo structural alerts | -- SYNTHESIS: Integrated Risk Assessment Risk classification, evidence summary, data gaps, recommendationsSeephase-procedures-detailed.mdfor complete tool parameters, decision logic, output templates, and fallback strategies for each phase.Tool Summary by PhasePhase 0: Compound DisambiguationPubChem_get_CID_by_compound_name(name: str)PubChem_get_compound_properties_by_CID(cid: int)ChEMBL_get_molecule(if ChEMBL ID available)Phase 1: Predictive ToxicologyDependency: ADMET-AI tools requirepip install tooluniverse[ml]. If unavailable, skip to Phase 3 and use CTD PubChemTox as alternatives.ADMETAI_predict_toxicity(smiles: list[str]) - AMES, DILI, ClinTox, LD50, hERG, etc.ADMETAI_predict_stress_response(smiles: list[str])ADMETAI_predict_nuclear_receptor_activity(smiles: list[str])Phase 2: ADMET PropertiesADMETAI_predict_BBB_penetrance/_bioavailability/_clearance_distribution/_CYP_interactions/_physicochemical_properties/_solubility_lipophilicity_hydration(all takesmiles: list[str])Phase 3: ToxicogenomicsCTD_get_chemical_gene_interactions(input_terms: str) — chemical name, returns gene interactions across speciesCTD_get_chemical_diseases(input_terms: str) — chemical-disease associations with evidence typePhase 3.5: PubChem Toxicity DataPubChemTox_get_toxicity_values(cid: int) — LD50, LC50, NOAEL reference valuesPubChemTox_get_ghs_classification(cid: int) — GHS hazard classification and pictogramsPubChemTox_get_carcinogen_classification(cid: int) — NTP/IARC carcinogenicity assessmentsPubChemTox_get_acute_effects(cid: int) — acute toxicity by route/speciesPubChemTox_get_toxicity_summary(cid: int) — integrated toxicity overviewPhase 3.6: Adverse Outcome PathwaysAOPWiki_list_aops(keyword: str) — search for relevant AOPs by chemical/mechanismAOPWiki_get_aop(aop_id: int) — full AOP detail: MIE, key events, adverse outcomePhase 4: Regulatory Safety (for pharmaceuticals only)Environmental chemicals: Skip Phases 4-5 (no FDA labels/DrugBank). Use CTD PubChemTox AOPWiki instead.FDA_get_boxed_warning_info_by_drug_name/_contraindications_/_adverse_reactions_/_warnings_(all takedrug_name: str)Phase 5: Drug Safety (for pharmaceuticals only)drugbank_get_safety_by_drug_name_or_drugbank_id(query,case_sensitive,exact_match,limit- all 4 required)Phase 6: Chemical-Protein InteractionsSTITCH_get_chemical_protein_interactions(identifiers: list[str],species: int)Fallback(if STITCH fails for industrial chemicals):STRING_get_interaction_partnersfor key target genes (e.g., ESR1 for endocrine disruptors)DGIdb_get_drug_gene_interactions(genes: list[str]) — for target druggability contextPhase 7: Structural AlertsChEMBL_search_compound_structural_alerts(molecule_chembl_id: str)Risk Classification MatrixRisk LevelCriteriaCRITICALFDA boxed warning OR multiple [T1] toxicity findings OR active DILI active hERGHIGHFDA warnings OR [T2] animal toxicity OR multiple active ADMET endpointsMEDIUMSome [T3] predictions positive OR CTD disease associations OR structural alertsLOWAll ADMET endpoints negative AND no FDA/DrugBank flags AND no CTD concernsINSUFFICIENT DATAFewer than 3 phases returned dataReport Structure# Chemical Safety Toxicology Report: [Compound Name] **Generated**: YYYY-MM-DD | **SMILES**: [...] | **CID**: [...] ## Executive Summary (risk classification key findings, all graded) ## 1. Compound Identity (disambiguation table) ## 2. Predictive Toxicology (ADMET-AI endpoints) ## 3. ADMET Profile (absorption, distribution, metabolism, excretion) ## 4. Toxicogenomics (CTD chemical-gene-disease) ## 5. Regulatory Safety (FDA label data) ## 6. Drug Safety Profile (DrugBank) ## 7. Chemical-Protein Interactions (STITCH network) ## 8. Structural Alerts (ChEMBL) ## 9. Integrated Risk Assessment (classification, evidence summary, gaps, recommendations) ## Appendix: Methods and Data SourcesSeereport-templates.mdfor full section templates with example tables.Mandatory Completeness ChecklistPhase 0: Compound disambiguated (SMILES CID minimum)Phase 1: At least 5 toxicity endpoints or “prediction unavailable”Phase 2: ADMET A/D/M/E sections or “not available”Phase 3: CTD queried; results or “no data in CTD”Phase 4: FDA labels queried; results or “not FDA-approved”Phase 5: DrugBank queried; results or “not found”Phase 6: STITCH queried; results or “no data available”Phase 7: Structural alerts checked or “ChEMBL ID not available”Synthesis: Risk classification with evidence summaryEvidence Grading: All findings have [T1]-[T4] annotationsData Gaps: Explicitly listedCommon Use PatternsNovel Compound: SMILES - Phase 0 (resolve) - Phase 1 (toxicity) - Phase 2 (ADMET) - Phase 7 (structural alerts) - SynthesisApproved Drug Review: Drug name - All phases (0-7) - Complete safety dossierEnvironmental Chemical: Chemical name - Phase 0 - Phase 1-2 - Phase 3 (CTD, key) - Phase 6 (STITCH) - SynthesisBatch Screening: Multiple SMILES - Phase 0 - Phase 1-2 (batch) - Comparative table - SynthesisToxicogenomic Deep-Dive: Chemical gene/disease interest - Phase 0 - Phase 3 (expanded CTD) - Literature - SynthesisLimitationsADMET-AI: Computational [T3]; should not replace experimental testingCTD: May lag behind latest literature by 6-12 monthsFDA: Only covers FDA-approved drugs; not applicable to environmental chemicalsDrugBank: Primarily drugs; limited industrial chemical coverageSTITCH: Lower score thresholds increase false positivesChEMBL: Structural alerts require ChEMBL ID; not all compounds have oneNovel compounds: May only have ADMET-AI predictions (no database evidence)SMILES validity: Invalid SMILES cause ADMET-AI failuresReference Filesphase-procedures-detailed.md- Complete tool parameters, decision logic, output templates, fallback strategies per phaseevidence-grading.md- Evidence grading details and examplesreport-templates.md- Full report section templates with example tablesphase-details.md- Additional phase contexttest_skill.py- Test suiteSummaryTotal tools integrated: 25 tools across 6 databases (ADMET-AI, CTD, FDA, DrugBank, STITCH, ChEMBL)Best for: Drug safety assessment, chemical hazard profiling, environmental toxicology, ADMET characterization, toxicogenomic analysisOutputs: Structured markdown report with risk classification (Critical/High/Medium/Low), evidence grading [T1-T4], and actionable recommendations