[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6DAS7AZZ28M8383CAAPM2QF":3,"skill-same:01M6DAS7AZZ28M8383CAAPM2QF":92},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":39,"linkOnly":40,"markdownContent":41,"markdownBody":42,"frontMatter":43,"refName":44,"commitSha":45,"skillDirectory":46,"forksCount":14,"repositoryHtmlUrl":47,"repositoryHomepageUrl":6,"ownerHtmlUrl":48,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":49,"install":50,"resources":63},"01M6DAS7AZZ28M8383CAAPM2QF","tooluniverse-immunotherapy-response-prediction",null,"Predict patient response to immune checkpoint inhibitors (ICIs) using multi-biomarker integration. Given a cancer type, somatic mutations, and optional biomarkers (TMB, PD-L1, MSI status), performs systematic analysis across 11 phases covering TMB classification, neoantigen burden estimation, MSI\u002FMMR assessment, PD-L1 evaluation, immune microenvironment profiling, mutation-based resistance\u002Fsensitivity prediction, clinical evidence retrieval, and multi-biomarker score integration. Generates a quantitative ICI Response Score (0-100), response likelihood tier, specific ICI drug recommendations with evidence, resistance risk factors, and a monitoring plan. Use when oncologists ask about immunotherapy eligibility, checkpoint inhibitor selection, or biomarker-guided ICI treatment decisions.","Prekzursil\u002Fagent-skills-toolchain","Prekzursil","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F54636077?v=4","01M6DACZ1PCQHD99SCR1YXG28B","Andrei Mihai Visalon","01M6DACZ670WZF535V3WYYSS4Z",0,"MIT","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Flibrary\u002Fskills\u002Ftooluniverse-immunotherapy-response-prediction\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"tooluniverse immunotherapy response prediction","evidence","patient","clinical","mutation","sensitivity","estimation","pd","selection","Immune","monitoring plan","use","decisions","ask","checkpoint","classification","analysis","risk","monitoring","assessment","2026-11-10 10:12:30",false,"---\nname: tooluniverse-immunotherapy-response-prediction\ndescription: Predict patient response to immune checkpoint inhibitors (ICIs) using multi-biomarker integration. Given a cancer type, somatic mutations, and optional biomarkers (TMB, PD-L1, MSI status), performs systematic analysis across 11 phases covering TMB classification, neoantigen burden estimation, MSI\u002FMMR assessment, PD-L1 evaluation, immune microenvironment profiling, mutation-based resistance\u002Fsensitivity prediction, clinical evidence retrieval, and multi-biomarker score integration. Generates a quantitative ICI Response Score (0-100), response likelihood tier, specific ICI drug recommendations with evidence, resistance risk factors, and a monitoring plan. Use when oncologists ask about immunotherapy eligibility, checkpoint inhibitor selection, or biomarker-guided ICI treatment decisions.\n---\n\n# Immunotherapy Response Prediction\n\nPredict patient response to immune checkpoint inhibitors (ICIs) using multi-biomarker integration. Transforms a patient tumor profile (cancer type + mutations + biomarkers) into a quantitative ICI Response Score with drug-specific recommendations, resistance risk assessment, and monitoring plan.\n\n## Reasoning Before Searching\n\nNot all tumors respond to checkpoint inhibitors. Reason through the biology before running tools:\n\n- **TMB (tumor mutational burden)**: More somatic mutations produce more neoantigens, which are recognized by T cells. High TMB (>=10 mut\u002FMb, FDA-approved threshold for pembrolizumab) generally predicts better response — but this varies by cancer type (e.g., RCC responds despite low TMB).\n- **MSI-H (microsatellite instability-high)**: Caused by defective DNA mismatch repair (MMR). MSI-H tumors have very high TMB and are pan-cancer approved for pembrolizumab. Check MLH1, MSH2, MSH6, PMS2 mutations.\n- **PD-L1 expression**: The direct target of pembrolizumab\u002Fatezolizumab. High PD-L1 (TPS >=50% or CPS >=10 depending on cancer) predicts response in some cancers (NSCLC) but not all (melanoma, where TMB is more predictive).\n- **Resistance factors** are equally important: STK11, KEAP1, JAK1\u002F2 loss, B2M mutations can render an otherwise TMB-high tumor non-responsive.\n\nBefore calling any tool, determine which biomarkers are available for this patient and which are unknown. This determines which phases can be scored with data vs. must use cancer-type priors. Do not default to \"moderate\" for unknowns — flag them explicitly as missing.\n\n**LOOK UP DON'T GUESS**: Never assume FDA approval for a biomarker-ICI combination — always verify with `fda_pharmacogenomic_biomarkers` or `FDA_get_indications_by_drug_name`. Cancer-specific thresholds differ from pan-cancer approvals.\n\n**KEY PRINCIPLES**:\n1. **Report-first approach** - Create report file FIRST, then populate progressively\n2. **Evidence-graded** - Every finding has an evidence tier (T1-T4)\n3. **Quantitative output** - ICI Response Score (0-100) with transparent component breakdown\n4. **Cancer-specific** - All thresholds and predictions are cancer-type adjusted\n5. **Multi-biomarker** - Integrate TMB + MSI + PD-L1 + neoantigen + mutations\n6. **Resistance-aware** - Always check for known resistance mutations (STK11, PTEN, JAK1\u002F2, B2M)\n7. **Drug-specific** - Recommend specific ICI agents with evidence\n8. **Source-referenced** - Every statement cites the tool\u002Fdatabase source\n9. **English-first queries** - Always use English terms in tool calls\n\n---\n\n## COMPUTE, DON'T DESCRIBE\nWhen 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.\n\n## When to Use\n\nApply when user asks:\n- \"Will this patient respond to immunotherapy?\"\n- \"Should I give pembrolizumab to this melanoma patient?\"\n- \"Patient has NSCLC with TMB 25, PD-L1 80% - predict ICI response\"\n- \"MSI-high colorectal cancer - which checkpoint inhibitor?\"\n- \"Patient has BRAF V600E melanoma, TMB 15 - immunotherapy or targeted?\"\n- \"Compare pembrolizumab vs nivolumab for this patient profile\"\n\n---\n\n## Input Parsing\n\n**Required**: Cancer type + at least one of: mutation list OR TMB value\n**Optional**: PD-L1 expression, MSI status, immune infiltration data, HLA type, prior treatments, intended ICI\n\nSee [INPUT_REFERENCE.md](INPUT_REFERENCE.md) for input format examples, cancer type normalization, and gene symbol normalization tables.\n\n---\n\n## Workflow Overview\n\n```\nInput: Cancer type + Mutations\u002FTMB + Optional biomarkers (PD-L1, MSI, etc.)\n\nPhase 1: Input Standardization & Cancer Context\nPhase 2: TMB Analysis\nPhase 3: Neoantigen Analysis\nPhase 4: MSI\u002FMMR Status Assessment\nPhase 5: PD-L1 Expression Analysis\nPhase 6: Immune Microenvironment Profiling\nPhase 7: Mutation-Based Predictors\nPhase 8: Clinical Evidence & ICI Options\nPhase 9: Resistance Risk Assessment\nPhase 10: Multi-Biomarker Score Integration\nPhase 11: Clinical Recommendations\n```\n\n---\n\n## Phase 1: Input Standardization & Cancer Context\n\n1. **Resolve cancer type** to EFO ID via `OpenTargets_get_disease_id_description_by_name`\n2. **Parse mutations** into structured format: `{gene, variant, type}`\n3. **Resolve gene IDs** via `MyGene_query_genes`\n4. Look up cancer-specific ICI baseline ORR from the cancer context table (see [SCORING_TABLES.md](SCORING_TABLES.md))\n\n## Phase 2: TMB Analysis\n\n1. Classify TMB: Very-Low (\u003C5), Low (5-9.9), Intermediate (10-19.9), High (>=20)\n2. Check FDA TMB-H biomarker via `fda_pharmacogenomic_biomarkers(drug_name='pembrolizumab')`\n3. Apply cancer-specific TMB thresholds (see [SCORING_TABLES.md](SCORING_TABLES.md))\n4. Note: RCC responds to ICIs despite low TMB; TMB is less predictive in some cancers\n\n## Phase 3: Neoantigen Analysis\n\n1. Estimate neoantigen burden: missense_count * 0.3 + frameshift_count * 1.5\n2. Check mutation impact via `UniProt_get_function_by_accession`\n3. Query known epitopes via `iedb_search_epitopes`\n4. POLE\u002FPOLD1 mutations indicate ultra-high neoantigen load\n\n## Phase 4: MSI\u002FMMR Status Assessment\n\n1. Integrate MSI status if provided (MSI-H = 25 pts, MSS = 5 pts)\n2. Check mutations in MMR genes: MLH1, MSH2, MSH6, PMS2, EPCAM\n3. Check FDA MSI-H approvals via `fda_pharmacogenomic_biomarkers(biomarker='Microsatellite Instability')`\n\n## Phase 5: PD-L1 Expression Analysis\n\n1. Classify PD-L1: High (>=50%), Positive (1-49%), Negative (\u003C1%)\n2. Apply cancer-specific PD-L1 thresholds and scoring methods (TPS vs CPS)\n3. Get baseline expression via `HPA_get_cancer_prognostics_by_gene(gene_name='CD274')`\n\n## Phase 6: Immune Microenvironment Profiling\n\n1. Query immune checkpoint gene expression for: CD274, PDCD1, CTLA4, LAG3, HAVCR2, TIGIT, CD8A, CD8B, GZMA, GZMB, PRF1, IFNG\n2. Classify tumor: Hot (T cell inflamed), Cold (immune desert), Immune excluded, Immune suppressed\n3. Run immune pathway enrichment via `enrichr_gene_enrichment_analysis`\n\n## Phase 7: Mutation-Based Predictors\n\n1. **Resistance mutations** (apply PENALTIES): STK11 (-10), PTEN (-5), JAK1\u002F2 (-10 each), B2M (-15), KEAP1 (-5), MDM2\u002F4 (-5), EGFR (-5)\n2. **Sensitivity mutations** (apply BONUSES): POLE (+10), POLD1 (+5), BRCA1\u002F2 (+3), ARID1A (+3), PBRM1 (+5 RCC only)\n3. Check CIViC and OpenTargets for driver mutation ICI context\n4. Check DDR pathway genes: ATM, ATR, CHEK1\u002F2, BRCA1\u002F2, PALB2, RAD50, MRE11\n\n## Phase 8: Clinical Evidence & ICI Options\n\n1. Query FDA indications for ICI drugs via `FDA_get_indications_by_drug_name`\n2. Search clinical trials via `search_clinical_trials` (params: `condition`, `intervention`, `query_term`)\n3. Search PubMed for biomarker-specific response data\n4. Get drug mechanisms via `OpenTargets_get_drug_mechanisms_of_action_by_chemblId`\n\nSee [SCORING_TABLES.md](SCORING_TABLES.md) for ICI drug profiles and ChEMBL IDs.\n\n## Phase 9: Resistance Risk Assessment\n\n1. Check CIViC for resistance evidence via `civic_search_evidence_items`\n2. Assess pathway-level resistance: IFN-g signaling, antigen presentation, WNT\u002Fb-catenin, MAPK, PI3K\u002FAKT\u002FmTOR\n3. Summarize risk: Low \u002F Moderate \u002F High\n\n## Phase 10: Multi-Biomarker Score Integration\n\n```\nTOTAL SCORE = TMB_score + MSI_score + PDL1_score + Neoantigen_score + Mutation_bonus + Resistance_penalty\n\nTMB_score:        5-30 points     MSI_score:        5-25 points\nPDL1_score:       5-20 points     Neoantigen_score: 5-15 points\nMutation_bonus:   0-10 points     Resistance_penalty: -20 to 0 points\n\nFloor: 0, Cap: 100\n```\n\n**Response Likelihood Tiers**:\n- 70-100 HIGH (50-80% ORR): Strong ICI candidate\n- 40-69 MODERATE (20-50% ORR): Consider ICI, combo preferred\n- 0-39 LOW (\u003C20% ORR): ICI alone unlikely effective\n\n**Confidence**: HIGH (all 4 biomarkers), MODERATE-HIGH (3\u002F4), MODERATE (2\u002F4), LOW (1), VERY LOW (cancer only)\n\n## Phase 11: Clinical Recommendations\n\n1. **ICI drug selection** using cancer-specific algorithm (see [SCORING_TABLES.md](SCORING_TABLES.md))\n2. **Monitoring plan**: CT\u002FMRI q8-12wk, ctDNA at 4-6wk, thyroid\u002Fliver function, irAEs\n3. **Alternative strategies** if LOW response: targeted therapy, chemotherapy, ICI+chemo combo, ICI+anti-angiogenic, ICI+CTLA-4 combo, clinical trials\n\n---\n\n## Output Report\n\nSave as `immunotherapy_response_prediction_{cancer_type}.md`. See [REPORT_TEMPLATE.md](REPORT_TEMPLATE.md) for the full report structure.\n\n---\n\n## Tool Parameter Reference\n\n**BEFORE calling ANY tool**, verify parameters. See [TOOLS_REFERENCE.md](TOOLS_REFERENCE.md) for verified tool parameters table.\n\nKey reminders:\n- `MyGene_query_genes`: use `query` (NOT `q`)\n- `EnsemblVEP_annotate_rsid`: use `variant_id` (NOT `rsid`)\n- `drugbank_*` tools: ALL 4 params required (`query`, `case_sensitive`, `exact_match`, `limit`)\n- `cBioPortal_get_mutations`: `gene_list` is a STRING not array\n- `ensembl_lookup_gene`: REQUIRES `species='homo_sapiens'`\n\n---\n\n## Evidence Tiers\n\n| Tier | Description | Source Examples |\n|------|-------------|----------------|\n| T1 | FDA-approved biomarker\u002Findication | FDA labels, NCCN guidelines |\n| T2 | Phase 2-3 clinical trial evidence | Published trial data, PubMed |\n| T3 | Preclinical\u002Fcomputational evidence | Pathway analysis, in vitro data |\n| T4 | Expert opinion\u002Fcase reports | Case series, reviews |\n\n---\n\n## References\n\n- OpenTargets: https:\u002F\u002Fplatform.opentargets.org\n- CIViC: https:\u002F\u002Fcivicdb.org\n- FDA Drug Labels: https:\u002F\u002Fdailymed.nlm.nih.gov\n- DrugBank: https:\u002F\u002Fgo.drugbank.com\n- PubMed: https:\u002F\u002Fpubmed.ncbi.nlm.nih.gov\n- IEDB: https:\u002F\u002Fwww.iedb.org\n- HPA: https:\u002F\u002Fwww.proteinatlas.org\n- cBioPortal: https:\u002F\u002Fwww.cbioportal.org\n","# Immunotherapy Response Prediction\n\nPredict patient response to immune checkpoint inhibitors (ICIs) using multi-biomarker integration. Transforms a patient tumor profile (cancer type + mutations + biomarkers) into a quantitative ICI Response Score with drug-specific recommendations, resistance risk assessment, and monitoring plan.\n\n## Reasoning Before Searching\n\nNot all tumors respond to checkpoint inhibitors. Reason through the biology before running tools:\n\n- **TMB (tumor mutational burden)**: More somatic mutations produce more neoantigens, which are recognized by T cells. High TMB (>=10 mut\u002FMb, FDA-approved threshold for pembrolizumab) generally predicts better response — but this varies by cancer type (e.g., RCC responds despite low TMB).\n- **MSI-H (microsatellite instability-high)**: Caused by defective DNA mismatch repair (MMR). MSI-H tumors have very high TMB and are pan-cancer approved for pembrolizumab. Check MLH1, MSH2, MSH6, PMS2 mutations.\n- **PD-L1 expression**: The direct target of pembrolizumab\u002Fatezolizumab. High PD-L1 (TPS >=50% or CPS >=10 depending on cancer) predicts response in some cancers (NSCLC) but not all (melanoma, where TMB is more predictive).\n- **Resistance factors** are equally important: STK11, KEAP1, JAK1\u002F2 loss, B2M mutations can render an otherwise TMB-high tumor non-responsive.\n\nBefore calling any tool, determine which biomarkers are available for this patient and which are unknown. This determines which phases can be scored with data vs. must use cancer-type priors. Do not default to \"moderate\" for unknowns — flag them explicitly as missing.\n\n**LOOK UP DON'T GUESS**: Never assume FDA approval for a biomarker-ICI combination — always verify with `fda_pharmacogenomic_biomarkers` or `FDA_get_indications_by_drug_name`. Cancer-specific thresholds differ from pan-cancer approvals.\n\n**KEY PRINCIPLES**:\n1. **Report-first approach** - Create report file FIRST, then populate progressively\n2. **Evidence-graded** - Every finding has an evidence tier (T1-T4)\n3. **Quantitative output** - ICI Response Score (0-100) with transparent component breakdown\n4. **Cancer-specific** - All thresholds and predictions are cancer-type adjusted\n5. **Multi-biomarker** - Integrate TMB + MSI + PD-L1 + neoantigen + mutations\n6. **Resistance-aware** - Always check for known resistance mutations (STK11, PTEN, JAK1\u002F2, B2M)\n7. **Drug-specific** - Recommend specific ICI agents with evidence\n8. **Source-referenced** - Every statement cites the tool\u002Fdatabase source\n9. **English-first queries** - Always use English terms in tool calls\n\n---\n\n## COMPUTE, DON'T DESCRIBE\nWhen 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.\n\n## When to Use\n\nApply when user asks:\n- \"Will this patient respond to immunotherapy?\"\n- \"Should I give pembrolizumab to this melanoma patient?\"\n- \"Patient has NSCLC with TMB 25, PD-L1 80% - predict ICI response\"\n- \"MSI-high colorectal cancer - which checkpoint inhibitor?\"\n- \"Patient has BRAF V600E melanoma, TMB 15 - immunotherapy or targeted?\"\n- \"Compare pembrolizumab vs nivolumab for this patient profile\"\n\n---\n\n## Input Parsing\n\n**Required**: Cancer type + at least one of: mutation list OR TMB value\n**Optional**: PD-L1 expression, MSI status, immune infiltration data, HLA type, prior treatments, intended ICI\n\nSee [INPUT_REFERENCE.md](INPUT_REFERENCE.md) for input format examples, cancer type normalization, and gene symbol normalization tables.\n\n---\n\n## Workflow Overview\n\n```\nInput: Cancer type + Mutations\u002FTMB + Optional biomarkers (PD-L1, MSI, etc.)\n\nPhase 1: Input Standardization & Cancer Context\nPhase 2: TMB Analysis\nPhase 3: Neoantigen Analysis\nPhase 4: MSI\u002FMMR Status Assessment\nPhase 5: PD-L1 Expression Analysis\nPhase 6: Immune Microenvironment Profiling\nPhase 7: Mutation-Based Predictors\nPhase 8: Clinical Evidence & ICI Options\nPhase 9: Resistance Risk Assessment\nPhase 10: Multi-Biomarker Score Integration\nPhase 11: Clinical Recommendations\n```\n\n---\n\n## Phase 1: Input Standardization & Cancer Context\n\n1. **Resolve cancer type** to EFO ID via `OpenTargets_get_disease_id_description_by_name`\n2. **Parse mutations** into structured format: `{gene, variant, type}`\n3. **Resolve gene IDs** via `MyGene_query_genes`\n4. Look up cancer-specific ICI baseline ORR from the cancer context table (see [SCORING_TABLES.md](SCORING_TABLES.md))\n\n## Phase 2: TMB Analysis\n\n1. Classify TMB: Very-Low (\u003C5), Low (5-9.9), Intermediate (10-19.9), High (>=20)\n2. Check FDA TMB-H biomarker via `fda_pharmacogenomic_biomarkers(drug_name='pembrolizumab')`\n3. Apply cancer-specific TMB thresholds (see [SCORING_TABLES.md](SCORING_TABLES.md))\n4. Note: RCC responds to ICIs despite low TMB; TMB is less predictive in some cancers\n\n## Phase 3: Neoantigen Analysis\n\n1. Estimate neoantigen burden: missense_count * 0.3 + frameshift_count * 1.5\n2. Check mutation impact via `UniProt_get_function_by_accession`\n3. Query known epitopes via `iedb_search_epitopes`\n4. POLE\u002FPOLD1 mutations indicate ultra-high neoantigen load\n\n## Phase 4: MSI\u002FMMR Status Assessment\n\n1. Integrate MSI status if provided (MSI-H = 25 pts, MSS = 5 pts)\n2. Check mutations in MMR genes: MLH1, MSH2, MSH6, PMS2, EPCAM\n3. Check FDA MSI-H approvals via `fda_pharmacogenomic_biomarkers(biomarker='Microsatellite Instability')`\n\n## Phase 5: PD-L1 Expression Analysis\n\n1. Classify PD-L1: High (>=50%), Positive (1-49%), Negative (\u003C1%)\n2. Apply cancer-specific PD-L1 thresholds and scoring methods (TPS vs CPS)\n3. Get baseline expression via `HPA_get_cancer_prognostics_by_gene(gene_name='CD274')`\n\n## Phase 6: Immune Microenvironment Profiling\n\n1. Query immune checkpoint gene expression for: CD274, PDCD1, CTLA4, LAG3, HAVCR2, TIGIT, CD8A, CD8B, GZMA, GZMB, PRF1, IFNG\n2. Classify tumor: Hot (T cell inflamed), Cold (immune desert), Immune excluded, Immune suppressed\n3. Run immune pathway enrichment via `enrichr_gene_enrichment_analysis`\n\n## Phase 7: Mutation-Based Predictors\n\n1. **Resistance mutations** (apply PENALTIES): STK11 (-10), PTEN (-5), JAK1\u002F2 (-10 each), B2M (-15), KEAP1 (-5), MDM2\u002F4 (-5), EGFR (-5)\n2. **Sensitivity mutations** (apply BONUSES): POLE (+10), POLD1 (+5), BRCA1\u002F2 (+3), ARID1A (+3), PBRM1 (+5 RCC only)\n3. Check CIViC and OpenTargets for driver mutation ICI context\n4. Check DDR pathway genes: ATM, ATR, CHEK1\u002F2, BRCA1\u002F2, PALB2, RAD50, MRE11\n\n## Phase 8: Clinical Evidence & ICI Options\n\n1. Query FDA indications for ICI drugs via `FDA_get_indications_by_drug_name`\n2. Search clinical trials via `search_clinical_trials` (params: `condition`, `intervention`, `query_term`)\n3. Search PubMed for biomarker-specific response data\n4. Get drug mechanisms via `OpenTargets_get_drug_mechanisms_of_action_by_chemblId`\n\nSee [SCORING_TABLES.md](SCORING_TABLES.md) for ICI drug profiles and ChEMBL IDs.\n\n## Phase 9: Resistance Risk Assessment\n\n1. Check CIViC for resistance evidence via `civic_search_evidence_items`\n2. Assess pathway-level resistance: IFN-g signaling, antigen presentation, WNT\u002Fb-catenin, MAPK, PI3K\u002FAKT\u002FmTOR\n3. Summarize risk: Low \u002F Moderate \u002F High\n\n## Phase 10: Multi-Biomarker Score Integration\n\n```\nTOTAL SCORE = TMB_score + MSI_score + PDL1_score + Neoantigen_score + Mutation_bonus + Resistance_penalty\n\nTMB_score:        5-30 points     MSI_score:        5-25 points\nPDL1_score:       5-20 points     Neoantigen_score: 5-15 points\nMutation_bonus:   0-10 points     Resistance_penalty: -20 to 0 points\n\nFloor: 0, Cap: 100\n```\n\n**Response Likelihood Tiers**:\n- 70-100 HIGH (50-80% ORR): Strong ICI candidate\n- 40-69 MODERATE (20-50% ORR): Consider ICI, combo preferred\n- 0-39 LOW (\u003C20% ORR): ICI alone unlikely effective\n\n**Confidence**: HIGH (all 4 biomarkers), MODERATE-HIGH (3\u002F4), MODERATE (2\u002F4), LOW (1), VERY LOW (cancer only)\n\n## Phase 11: Clinical Recommendations\n\n1. **ICI drug selection** using cancer-specific algorithm (see [SCORING_TABLES.md](SCORING_TABLES.md))\n2. **Monitoring plan**: CT\u002FMRI q8-12wk, ctDNA at 4-6wk, thyroid\u002Fliver function, irAEs\n3. **Alternative strategies** if LOW response: targeted therapy, chemotherapy, ICI+chemo combo, ICI+anti-angiogenic, ICI+CTLA-4 combo, clinical trials\n\n---\n\n## Output Report\n\nSave as `immunotherapy_response_prediction_{cancer_type}.md`. See [REPORT_TEMPLATE.md](REPORT_TEMPLATE.md) for the full report structure.\n\n---\n\n## Tool Parameter Reference\n\n**BEFORE calling ANY tool**, verify parameters. See [TOOLS_REFERENCE.md](TOOLS_REFERENCE.md) for verified tool parameters table.\n\nKey reminders:\n- `MyGene_query_genes`: use `query` (NOT `q`)\n- `EnsemblVEP_annotate_rsid`: use `variant_id` (NOT `rsid`)\n- `drugbank_*` tools: ALL 4 params required (`query`, `case_sensitive`, `exact_match`, `limit`)\n- `cBioPortal_get_mutations`: `gene_list` is a STRING not array\n- `ensembl_lookup_gene`: REQUIRES `species='homo_sapiens'`\n\n---\n\n## Evidence Tiers\n\n| Tier | Description | Source Examples |\n|------|-------------|----------------|\n| T1 | FDA-approved biomarker\u002Findication | FDA labels, NCCN guidelines |\n| T2 | Phase 2-3 clinical trial evidence | Published trial data, PubMed |\n| T3 | Preclinical\u002Fcomputational evidence | Pathway analysis, in vitro data |\n| T4 | Expert opinion\u002Fcase reports | Case series, reviews |\n\n---\n\n## References\n\n- OpenTargets: https:\u002F\u002Fplatform.opentargets.org\n- CIViC: https:\u002F\u002Fcivicdb.org\n- FDA Drug Labels: https:\u002F\u002Fdailymed.nlm.nih.gov\n- DrugBank: https:\u002F\u002Fgo.drugbank.com\n- PubMed: https:\u002F\u002Fpubmed.ncbi.nlm.nih.gov\n- IEDB: https:\u002F\u002Fwww.iedb.org\n- HPA: https:\u002F\u002Fwww.proteinatlas.org\n- cBioPortal: https:\u002F\u002Fwww.cbioportal.org\n",{"name":5,"description":7},"main","a298c929b6cbd662e1fce54302ce520d5ceea1a5",".agents\u002Flibrary\u002Fskills\u002Ftooluniverse-immunotherapy-response-prediction","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain","https:\u002F\u002Fgithub.com\u002FPrekzursil","2026-11-02 14:30:57",{"source":8,"skillName":5,"commands":51,"prompt":61,"zipAvailable":62},[52,55,58],{"tool":53,"command":54},"npx","npx skills add Prekzursil\u002Fagent-skills-toolchain --skill tooluniverse-immunotherapy-response-prediction",{"tool":56,"command":57},"bunx","bunx skills add Prekzursil\u002Fagent-skills-toolchain --skill tooluniverse-immunotherapy-response-prediction",{"tool":59,"command":60},"pnpm","pnpm dlx skills add Prekzursil\u002Fagent-skills-toolchain --skill tooluniverse-immunotherapy-response-prediction","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Flibrary\u002Fskills\u002Ftooluniverse-immunotherapy-response-prediction\u002FSKILL.md",true,[64,70,74,78,82,88],{"path":65,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":68,"sourceHtmlUrl":69},"EXAMPLES.md","REFERENCE","参考文档",4807,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Flibrary\u002Fskills\u002Ftooluniverse-immunotherapy-response-prediction\u002FEXAMPLES.md",{"path":71,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":72,"sourceHtmlUrl":73},"INPUT_REFERENCE.md",1585,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Flibrary\u002Fskills\u002Ftooluniverse-immunotherapy-response-prediction\u002FINPUT_REFERENCE.md",{"path":75,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":76,"sourceHtmlUrl":77},"REPORT_TEMPLATE.md",5146,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Flibrary\u002Fskills\u002Ftooluniverse-immunotherapy-response-prediction\u002FREPORT_TEMPLATE.md",{"path":79,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":80,"sourceHtmlUrl":81},"SCORING_TABLES.md",7341,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Flibrary\u002Fskills\u002Ftooluniverse-immunotherapy-response-prediction\u002FSCORING_TABLES.md",{"path":83,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":86,"sourceHtmlUrl":87},"test_skill.py","SCRIPT","脚本",47090,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Flibrary\u002Fskills\u002Ftooluniverse-immunotherapy-response-prediction\u002Ftest_skill.py",{"path":89,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":90,"sourceHtmlUrl":91},"TOOLS_REFERENCE.md",6406,"https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002F.agents\u002Flibrary\u002Fskills\u002Ftooluniverse-immunotherapy-response-prediction\u002FTOOLS_REFERENCE.md",[93,113,144,171,185,201],{"skillNo":94,"name":95,"nameZh":6,"description":96,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"publishedAt":39},"01M6DF7D7478R55P46REYH4QAC","clarify-requirements","Resolve genuine pre-execution ambiguity by scanning repo context, generating 1-6 grounded options, and asking 1-3 focused questions via AskUserQuestion. NOT for mid-task use (injection risk); NOT when the goal is already unambiguous; use to disambiguate vague prompts before any file edits.","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Fworkspace-hooks\u002Fskills\u002Fclarify-requirements\u002FSKILL.md",[],[100,101,102,36,103,104,105,106,107,30,108,109,110,111,112],"clarify requirements","clarify","requirements","options","task","prompts","goal","scanning","context","repo","injection","execution","file",{"skillNo":114,"name":115,"nameZh":6,"description":116,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":124,"publishedAt":39},"01M6DF79CA4BHJS66J8EZDBPWJ","openmm-md","Run FREE, open-source, Windows-native GPU molecular dynamics via the openmm-md MCP (OpenMM + CUDA on the local NVIDIA GPU). Use for protein \u002F protein-ligand MD when you want a free stack or are on Windows without a Schrodinger license. Covers the full workflow (prepare, solvate, minimize, equilibrate, production, analyze, mmgbsa) AND an honest decision guide vs Schrodinger Desmond, including how to run Desmond legitimately on Linux or a cloud GPU with a real license. This is a deliberately LESSER free alternative and does NOT replace the schrodinger MCP tools. Do NOT use it to pick a free tool when a validated FEP+, WaterMap, IFD-MD, or Prime MM-GBSA result is actually required.","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Fmcp\u002Fopenmm-md-mcp\u002Fskill\u002Fopenmm-md\u002FSKILL.md",[119],{"categoryNo":120,"categoryName":121,"taxonomyType":122,"taxonomyTypeDesc":123},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[125,126,127,128,129,130,131,132,133,134,135,136,137,138,30,139,140,141,142,143],"openmm md","how","decision","pick","linux","analyze","tool","stack","open","NVIDIA","gpu","license","workflow","windows","free","molecular dynamics","local","do","MCP",{"skillNo":145,"name":146,"nameZh":6,"description":147,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":148,"categories":149,"keywords":156,"publishedAt":39},"01M6DF76C1MG85VEWP8VZCBNMN","visual-review","Take screenshots of web pages and UI using Playwright for visual review and iteration","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fvisual-review\u002FSKILL.md",[150,153],{"categoryNo":151,"categoryName":152,"taxonomyType":122,"taxonomyTypeDesc":123},"01JBSEEDCAT000000000000003","演示文稿",{"categoryNo":154,"categoryName":155,"taxonomyType":122,"taxonomyTypeDesc":123},"01JBSEEDCAT000000000000008","测试与质量",[157,158,159,160,161,162,163,164,165,166,126,167,168,169,170],"review","visual","visual review","pages","web","false","iteration","take","check","slides","ui","screenshot","playwright","screenshots",{"skillNo":172,"name":173,"nameZh":6,"description":174,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":175,"categories":176,"keywords":177,"publishedAt":39},"01M6DF73AQVCBZKEN0YXEX9FXP","status","Diagnostic status report — shows metaswarm installation state, project setup, and potential issues","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fstatus\u002FSKILL.md",[],[173,178,179,180,181,182,183,184],"setup","report","project","installation","project setup","issues","state",{"skillNo":186,"name":187,"nameZh":6,"description":188,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":189,"categories":190,"keywords":192,"publishedAt":39},"01M6DF6X6C2XB0P6GX76P7TTYV","start","Use when starting work on any task, when the user mentions metaswarm, or when the user wants to begin tracked development work","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fstart\u002FSKILL.md",[191],{"categoryNo":120,"categoryName":121,"taxonomyType":122,"taxonomyTypeDesc":123},[187,193,104,194,195,196,30,197,198,199,200],"start issue","issue","ready","work on issue","user","work","work on","development",{"skillNo":202,"name":178,"nameZh":6,"description":203,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":204,"categories":205,"keywords":206,"publishedAt":39},"01M6DF6QMJ9AJKFKHHKRT005PQ","Interactive project setup — detects your project, configures metaswarm, writes project-local files","https:\u002F\u002Fgithub.com\u002FPrekzursil\u002Fagent-skills-toolchain\u002Fblob\u002Fmain\u002Ftools\u002Feditor\u002Fmetaswarm\u002Fskills\u002Fsetup\u002FSKILL.md",[],[178,180,141,207,182,208],"files","interactive"]