tooluniverse 1.0.7__py3-none-any.whl → 1.0.9__py3-none-any.whl

tooluniverse/data/compose_tools.json

@@ -314,59 +314,5 @@
     ],
     "composition_file": "tool_graph_generation.py",
     "composition_function": "compose"
-  },
-  {
-    "type": "ComposeTool",
-    "name": "MultiAgentLiteratureSearch",
-    "description": "Multi-agent literature search system that uses AI agents to analyze intent, extract keywords, execute parallel searches, summarize results, and check quality iteratively",
-    "parameter": {
-      "type": "object",
-      "properties": {
-        "query": {
-          "type": "string",
-          "description": "The research query to search for"
-        },
-        "max_iterations": {
-          "type": "integer",
-          "description": "Maximum number of iterations (default: 3)",
-          "default": 3
-        },
-        "quality_threshold": {
-          "type": "number",
-          "description": "Quality threshold for completion (default: 0.7)",
-          "default": 0.7
-        }
-      },
-      "required": [
-        "query",
-        "max_iterations",
-        "quality_threshold"
-      ]
-    },
-    "auto_load_dependencies": true,
-    "fail_on_missing_tools": false,
-    "required_tools": [
-      "IntentAnalyzerAgent",
-      "KeywordExtractorAgent",
-      "ResultSummarizerAgent",
-      "QualityCheckerAgent",
-      "OverallSummaryAgent",
-      "ArXiv_search_papers",
-      "BioRxiv_search_preprints",
-      "MedRxiv_search_preprints",
-      "HAL_search_archive",
-      "Crossref_search_works",
-      "PubMed_search_articles",
-      "EuropePMC_search_articles",
-      "SemanticScholar_search_papers",
-      "openalex_literature_search",
-      "DBLP_search_publications",
-      "DOAJ_search_articles",
-      "CORE_search_papers",
-      "PMC_search_papers",
-      "Zenodo_search_records"
-    ],
-    "composition_file": "multi_agent_literature_search.py",
-    "composition_function": "compose"
   }
 ]

tooluniverse/data/core_tools.json

@@ -32,10 +32,7 @@
       },
       "required": [
         "query",
-        "limit",
-        "year_from",
-        "year_to",
-        "language"
+        "limit"
       ]
     },
     "return_schema": {

tooluniverse/data/dataset_tools.json

@@ -7,7 +7,7 @@
       "repo_id": "agenticx/DrugbankVocabulary",
       "path_in_repo": "drugbank_vocabulary.csv",
       "token": null,
-      "save_to_local_dir": "src/tooluniverse/test"
+      "save_to_local_dir": null
     },
     "parameter": {
       "type": "object",
@@ -145,7 +145,7 @@
       "repo_id": "agenticx/DrugbankVocabulary",
       "path_in_repo": "drugbank_vocabulary.csv",
       "token": null,
-      "save_to_local_dir": "src/tooluniverse/test"
+      "save_to_local_dir": null
     },
     "parameter": {
       "type": "object",
@@ -281,7 +281,7 @@
       "repo_id": "agenticx/DrugbankRawParquet",
       "path_in_repo": "drugbank_raw.parquet",
       "token": null,
-      "save_to_local_dir": "src/tooluniverse/test"
+      "save_to_local_dir": null
     },
     "parameter": {
       "type": "object",
@@ -471,7 +471,7 @@
       "repo_id": "agenticx/DrugLinks",
       "path_in_repo": "drug_links.csv",
       "token": null,
-      "save_to_local_dir": "src/tooluniverse/test"
+      "save_to_local_dir": null
     },
     "parameter": {
       "type": "object",
@@ -707,7 +707,7 @@
       "repo_id": "agenticx/DICTrank",
       "path_in_repo": "DICTrank.csv",
       "token": null,
-      "save_to_local_dir": "src/tooluniverse/test"
+      "save_to_local_dir": null
     },
     "parameter": {
       "type": "object",
@@ -845,7 +845,7 @@
       "repo_id": "agenticx/DILIrank",
       "path_in_repo": "DILIrank.csv",
       "token": null,
-      "save_to_local_dir": "src/tooluniverse/test"
+      "save_to_local_dir": null
     },
     "parameter": {
       "type": "object",
@@ -963,7 +963,7 @@
       "repo_id": "agenticx/DIQTA",
       "path_in_repo": "DIQTA.csv",
       "token": null,
-      "save_to_local_dir": "src/tooluniverse/test"
+      "save_to_local_dir": null
     },
     "parameter": {
       "type": "object",

tooluniverse/data/doaj_tools.json

@@ -22,9 +22,7 @@
         }
       },
       "required": [
-        "query",
-        "max_results",
-        "type"
+        "query"
       ]
     },
     "return_schema": {

tooluniverse/data/drug_discovery_agents.json

@@ -0,0 +1,282 @@
+[
+  {
+    "type": "AgenticTool",
+    "name": "DiseaseAnalyzerAgent",
+    "description": "AI agent that analyzes disease characteristics and identifies potential therapeutic targets using LLM reasoning",
+    "prompt": "You are a biomedical expert specializing in disease analysis and target identification. Analyze the following disease and provide comprehensive insights:\n\nDisease: {disease_name}\nContext: {context}\n\nPlease provide:\n1. Disease Overview: Key characteristics, pathophysiology, and clinical manifestations\n2. Molecular Mechanisms: Underlying biological processes and pathways\n3. Target Identification: Potential therapeutic targets with rationale\n4. Target Prioritization: Rank targets by druggability and therapeutic potential\n5. Research Gaps: Areas needing further investigation\n\nReturn as structured JSON:\n{\n  \"disease_overview\": {\n    \"characteristics\": \"...\",\n    \"pathophysiology\": \"...\",\n    \"clinical_manifestations\": \"...\"\n  },\n  \"molecular_mechanisms\": [\n    {\n      \"pathway\": \"...\",\n      \"description\": \"...\",\n      \"relevance\": \"...\"\n    }\n  ],\n  \"therapeutic_targets\": [\n    {\n      \"target_name\": \"...\",\n      \"target_type\": \"...\",\n      \"rationale\": \"...\",\n      \"druggability_score\": 1-10,\n      \"therapeutic_potential\": 1-10\n    }\n  ],\n  \"research_gaps\": [\"...\", \"...\"]\n}",
+    "input_arguments": [
+      "disease_name",
+      "context"
+    ],
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "disease_name": {
+          "type": "string",
+          "description": "Name of the disease to analyze"
+        },
+        "context": {
+          "type": "string",
+          "description": "Additional context or specific focus areas",
+          "default": ""
+        }
+      },
+      "required": [
+        "disease_name"
+      ]
+    },
+    "configs": {
+      "api_type": "CHATGPT",
+      "model_id": "gpt-4.1",
+      "temperature": 0.7,
+      "max_new_tokens": 3000,
+      "return_json": true
+    }
+  },
+  {
+    "type": "AgenticTool",
+    "name": "CompoundDiscoveryAgent",
+    "description": "AI agent that analyzes potential drug compounds using multiple strategies and LLM reasoning",
+    "prompt": "You are a drug discovery expert. Based on the disease and targets provided, discover potential therapeutic compounds using multiple strategies:\n\nDisease: {disease_name}\nTargets: {targets}\nDiscovery Context: {context}\n\nPlease provide:\n1. Target-Based Discovery: Compounds that interact with specific targets\n2. Mechanism-Based Discovery: Compounds affecting disease mechanisms\n3. Repurposing Candidates: Existing drugs that could be repurposed\n4. Natural Products: Plant-derived or natural compounds\n5. Novel Compounds: Recently discovered or experimental compounds\n6. Discovery Rationale: Why each compound might be effective\n\nReturn as structured JSON:\n{\n  \"target_based_compounds\": [\n    {\n      \"compound_name\": \"...\",\n      \"target\": \"...\",\n      \"mechanism\": \"...\",\n      \"rationale\": \"...\",\n      \"confidence_score\": 1-10\n    }\n  ],\n  \"mechanism_based_compounds\": [\n    {\n      \"compound_name\": \"...\",\n      \"mechanism\": \"...\",\n      \"rationale\": \"...\",\n      \"confidence_score\": 1-10\n    }\n  ],\n  \"repurposing_candidates\": [\n    {\n      \"compound_name\": \"...\",\n      \"current_use\": \"...\",\n      \"repurposing_rationale\": \"...\",\n      \"confidence_score\": 1-10\n    }\n  ],\n  \"natural_products\": [\n    {\n      \"compound_name\": \"...\",\n      \"source\": \"...\",\n      \"mechanism\": \"...\",\n      \"rationale\": \"...\",\n      \"confidence_score\": 1-10\n    }\n  ],\n  \"novel_compounds\": [\n    {\n      \"compound_name\": \"...\",\n      \"discovery_status\": \"...\",\n      \"mechanism\": \"...\",\n      \"rationale\": \"...\",\n      \"confidence_score\": 1-10\n    }\n  ]\n}",
+    "input_arguments": [
+      "disease_name",
+      "targets",
+      "context"
+    ],
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "disease_name": {
+          "type": "string",
+          "description": "Name of the disease"
+        },
+        "targets": {
+          "type": "string",
+          "description": "List of therapeutic targets (comma-separated)"
+        },
+        "context": {
+          "type": "string",
+          "description": "Additional context or specific requirements",
+          "default": ""
+        }
+      },
+      "required": [
+        "disease_name",
+        "targets"
+      ]
+    },
+    "configs": {
+      "api_type": "CHATGPT",
+      "model_id": "gpt-4.1",
+      "temperature": 0.8,
+      "max_new_tokens": 4000,
+      "return_json": true
+    }
+  },
+  {
+    "type": "AgenticTool",
+    "name": "ADMETAnalyzerAgent",
+    "description": "AI agent that analyzes ADMET data and provides insights on drug-likeness and safety profiles",
+    "prompt": "You are a pharmaceutical expert specializing in ADMET (Absorption, Distribution, Metabolism, Excretion, Toxicity) analysis. Analyze the following ADMET data and provide insights:\n\nCompounds: {compounds}\nADMET Data: {admet_data}\nDisease Context: {disease_context}\n\nFor each compound, provide:\n1. Data Interpretation: What the ADMET data tells us\n2. Strengths: Positive ADMET characteristics\n3. Concerns: Potential ADMET issues or red flags\n4. Drug-likeness Assessment: Compliance with drug-like properties\n5. Safety Profile Analysis: Overall safety assessment based on data\n6. Comparative Analysis: How compounds compare to each other\n7. Clinical Implications: What this means for clinical development\n8. Optimization Opportunities: Areas for improvement based on data\n\nReturn as structured JSON:\n{\n  \"admet_analysis\": [\n    {\n      \"compound_name\": \"...\",\n      \"data_interpretation\": {\n        \"absorption_insights\": \"...\",\n        \"distribution_insights\": \"...\",\n        \"metabolism_insights\": \"...\",\n        \"excretion_insights\": \"...\",\n        \"toxicity_insights\": \"...\"\n      },\n      \"strengths\": [\"...\", \"...\"],\n      \"concerns\": [\"...\", \"...\"],\n      \"drug_likeness_assessment\": {\n        \"overall_compliance\": \"good/moderate/poor\",\n        \"specific_issues\": [\"...\", \"...\"],\n        \"recommendations\": [\"...\", \"...\"]\n      },\n      \"safety_profile\": {\n        \"overall_assessment\": \"safe/moderate_risk/high_risk\",\n        \"key_concerns\": [\"...\", \"...\"],\n        \"monitoring_needs\": [\"...\", \"...\"]\n      },\n      \"clinical_implications\": [\"...\", \"...\"],\n      \"optimization_opportunities\": [\"...\", \"...\"]\n    }\n  ],\n  \"comparative_analysis\": {\n    \"best_performing\": \"...\",\n    \"ranking_rationale\": \"...\",\n    \"overall_recommendations\": [\"...\", \"...\"]\n  }\n}",
+    "input_arguments": [
+      "compounds",
+      "admet_data",
+      "disease_context"
+    ],
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "compounds": {
+          "type": "string",
+          "description": "List of compounds to analyze (comma-separated)"
+        },
+        "admet_data": {
+          "type": "string",
+          "description": "ADMET data from computational tools to analyze"
+        },
+        "disease_context": {
+          "type": "string",
+          "description": "Disease context for ADMET evaluation",
+          "default": ""
+        }
+      },
+      "required": [
+        "compounds",
+        "admet_data"
+      ]
+    },
+    "configs": {
+      "api_type": "CHATGPT",
+      "model_id": "gpt-4.1",
+      "temperature": 0.6,
+      "max_new_tokens": 5000,
+      "return_json": true
+    }
+  },
+  {
+    "type": "AgenticTool",
+    "name": "DrugInteractionAnalyzerAgent",
+    "description": "AI agent that analyzes drug-drug interactions and provides clinical recommendations",
+    "prompt": "You are a clinical pharmacology expert. Analyze potential drug-drug interactions for the following compounds:\n\nCompounds: {compounds}\nPatient Context: {patient_context}\n\nFor each compound pair, provide:\n1. Interaction Type: Pharmacokinetic, pharmacodynamic, or both\n2. Mechanism: How the interaction occurs\n3. Severity: Minor, moderate, major, or contraindicated\n4. Clinical Significance: Impact on patient care\n5. Management Strategy: How to handle the interaction\n6. Monitoring Requirements: What to monitor\n7. Alternative Options: Safer alternatives if needed\n\nReturn as structured JSON:\n{\n  \"interaction_analysis\": [\n    {\n      \"compound_pair\": \"compound1 + compound2\",\n      \"interaction_type\": \"pharmacokinetic/pharmacodynamic/both\",\n      \"mechanism\": \"...\",\n      \"severity\": \"minor/moderate/major/contraindicated\",\n      \"clinical_significance\": \"...\",\n      \"management_strategy\": \"...\",\n      \"monitoring_requirements\": [\"...\", \"...\"],\n      \"alternative_options\": [\"...\", \"...\"]\n    }\n  ],\n  \"overall_risk_assessment\": {\n    \"risk_level\": \"low/medium/high\",\n    \"recommendations\": [\"...\", \"...\"],\n    \"monitoring_plan\": \"...\"\n  }\n}",
+    "input_arguments": [
+      "compounds",
+      "patient_context"
+    ],
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "compounds": {
+          "type": "string",
+          "description": "List of compounds to analyze for interactions (comma-separated)"
+        },
+        "patient_context": {
+          "type": "string",
+          "description": "Patient context (age, comorbidities, medications, etc.)",
+          "default": "General adult population"
+        }
+      },
+      "required": [
+        "compounds"
+      ]
+    },
+    "configs": {
+      "api_type": "CHATGPT",
+      "model_id": "gpt-4.1",
+      "temperature": 0.5,
+      "max_new_tokens": 3000,
+      "return_json": true
+    }
+  },
+  {
+    "type": "AgenticTool",
+    "name": "LiteratureSynthesisAgent",
+    "description": "AI agent that synthesizes literature findings and provides evidence-based insights",
+    "prompt": "You are a biomedical literature expert. Synthesize the following research findings and provide evidence-based insights:\n\nResearch Topic: {topic}\nLiterature Findings: {literature_data}\nFocus Area: {focus_area}\n\nPlease provide:\n1. Key Findings: Most important discoveries\n2. Evidence Quality: Assessment of evidence strength\n3. Consensus: Areas of agreement in the field\n4. Controversies: Areas of disagreement or uncertainty\n5. Gaps: Research gaps and future directions\n6. Clinical Relevance: Translation to clinical practice\n7. Recommendations: Evidence-based recommendations\n\nReturn as structured JSON:\n{\n  \"key_findings\": [\n    {\n      \"finding\": \"...\",\n      \"evidence_strength\": \"strong/moderate/weak\",\n      \"source_count\": number,\n      \"clinical_relevance\": \"high/medium/low\"\n    }\n  ],\n  \"evidence_quality\": {\n    \"overall_strength\": \"strong/moderate/weak\",\n    \"study_types\": [\"...\", \"...\"],\n    \"sample_sizes\": \"...\",\n    \"methodology_quality\": \"...\"\n  },\n  \"consensus_areas\": [\"...\", \"...\"],\n  \"controversies\": [\n    {\n      \"issue\": \"...\",\n      \"perspectives\": [\"...\", \"...\"],\n      \"resolution_status\": \"resolved/unresolved/ongoing\"\n    }\n  ],\n  \"research_gaps\": [\"...\", \"...\"],\n  \"clinical_relevance\": {\n    \"translation_potential\": \"high/medium/low\",\n    \"clinical_implications\": [\"...\", \"...\"],\n    \"implementation_barriers\": [\"...\", \"...\"]\n  },\n  \"recommendations\": [\n    {\n      \"recommendation\": \"...\",\n      \"evidence_level\": \"A/B/C/D\",\n      \"priority\": \"high/medium/low\"\n    }\n  ]\n}",
+    "input_arguments": [
+      "topic",
+      "literature_data",
+      "focus_area"
+    ],
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "topic": {
+          "type": "string",
+          "description": "Research topic or question"
+        },
+        "literature_data": {
+          "type": "string",
+          "description": "Literature findings or abstracts to synthesize"
+        },
+        "focus_area": {
+          "type": "string",
+          "description": "Specific focus area for synthesis",
+          "default": "General"
+        }
+      },
+      "required": [
+        "topic",
+        "literature_data"
+      ]
+    },
+    "configs": {
+      "api_type": "CHATGPT",
+      "model_id": "gpt-4.1",
+      "temperature": 0.6,
+      "max_new_tokens": 4000,
+      "return_json": true
+    }
+  },
+  {
+    "type": "AgenticTool",
+    "name": "DrugOptimizationAgent",
+    "description": "AI agent that analyzes drug optimization strategies based on ADMET and efficacy data",
+    "prompt": "You are a medicinal chemistry expert. Based on the provided data, analyze optimization strategies for the following compounds:\n\nCompounds: {compounds}\nADMET Data: {admet_data}\nEfficacy Data: {efficacy_data}\nTarget Profile: {target_profile}\n\nFor each compound, provide:\n1. Current Strengths: What works well\n2. Current Weaknesses: What needs improvement\n3. Optimization Strategies: Specific modifications to consider\n4. Synthetic Feasibility: How difficult to implement\n5. Expected Improvements: Potential outcomes based on data\n6. Risk Assessment: Potential risks of modifications\n7. Priority Ranking: Which optimizations to pursue first\n\nReturn as structured JSON:\n{\n  \"optimization_analysis\": [\n    {\n      \"compound_name\": \"...\",\n      \"current_strengths\": [\"...\", \"...\"],\n      \"current_weaknesses\": [\"...\", \"...\"],\n      \"optimization_strategies\": [\n        {\n          \"strategy\": \"...\",\n          \"modification\": \"...\",\n          \"rationale\": \"...\",\n          \"synthetic_feasibility\": \"easy/moderate/difficult\",\n          \"expected_improvement\": \"...\",\n          \"risk_level\": \"low/medium/high\",\n          \"priority\": 1-5\n        }\n      ]\n    }\n  ],\n  \"overall_recommendations\": {\n    \"top_priorities\": [\"...\", \"...\"],\n    \"synthesis_roadmap\": \"...\",\n    \"success_probability\": \"high/medium/low\"\n  }\n}",
+    "input_arguments": [
+      "compounds",
+      "admet_data",
+      "efficacy_data",
+      "target_profile"
+    ],
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "compounds": {
+          "type": "string",
+          "description": "List of compounds to optimize (comma-separated)"
+        },
+        "admet_data": {
+          "type": "string",
+          "description": "ADMET properties and issues",
+          "default": ""
+        },
+        "efficacy_data": {
+          "type": "string",
+          "description": "Efficacy and potency data",
+          "default": ""
+        },
+        "target_profile": {
+          "type": "string",
+          "description": "Target profile and requirements",
+          "default": ""
+        }
+      },
+      "required": [
+        "compounds"
+      ]
+    },
+    "configs": {
+      "api_type": "CHATGPT",
+      "model_id": "gpt-4.1",
+      "temperature": 0.7,
+      "max_new_tokens": 4000,
+      "return_json": true
+    }
+  },
+  {
+    "type": "AgenticTool",
+    "name": "ClinicalTrialDesignAgent",
+    "description": "AI agent that designs clinical trial protocols based on preclinical data and regulatory requirements",
+    "prompt": "You are a clinical trial design expert. Design a clinical trial protocol based on the following information:\n\nDrug: {drug_name}\nIndication: {indication}\nPreclinical Data: {preclinical_data}\nTarget Population: {target_population}\n\nPlease provide:\n1. Study Design: Phase, design type, duration\n2. Primary Endpoints: Main efficacy endpoints\n3. Secondary Endpoints: Additional endpoints\n4. Inclusion/Exclusion Criteria: Patient selection criteria\n5. Dosing Strategy: Dose selection and escalation\n6. Safety Monitoring: Safety endpoints and monitoring plan\n7. Statistical Plan: Sample size, power, analysis plan\n8. Regulatory Considerations: FDA/EMA requirements\n9. Timeline: Study duration and milestones\n10. Risk Mitigation: Strategies to address potential risks\n\nReturn as structured JSON:\n{\n  \"study_design\": {\n    \"phase\": \"I/II/III/IV\",\n    \"design_type\": \"...\",\n    \"duration\": \"...\",\n    \"randomization\": \"...\",\n    \"blinding\": \"...\"\n  },\n  \"endpoints\": {\n    \"primary\": [\n      {\n        \"endpoint\": \"...\",\n        \"measurement\": \"...\",\n        \"timeline\": \"...\"\n      }\n    ],\n    \"secondary\": [\n      {\n        \"endpoint\": \"...\",\n        \"measurement\": \"...\",\n        \"timeline\": \"...\"\n      }\n    ]\n  },\n  \"patient_criteria\": {\n    \"inclusion\": [\"...\", \"...\"],\n    \"exclusion\": [\"...\", \"...\"]\n  },\n  \"dosing_strategy\": {\n    \"starting_dose\": \"...\",\n    \"dose_escalation\": \"...\",\n    \"max_dose\": \"...\",\n    \"administration\": \"...\"\n  },\n  \"safety_monitoring\": {\n    \"safety_endpoints\": [\"...\", \"...\"],\n    \"monitoring_plan\": \"...\",\n    \"stopping_rules\": [\"...\", \"...\"]\n  },\n  \"statistical_plan\": {\n    \"sample_size\": \"...\",\n    \"power_calculation\": \"...\",\n    \"analysis_plan\": \"...\"\n  },\n  \"regulatory_considerations\": [\"...\", \"...\"],\n  \"timeline\": {\n    \"total_duration\": \"...\",\n    \"milestones\": [\"...\", \"...\"]\n  },\n  \"risk_mitigation\": [\"...\", \"...\"]\n}",
+    "input_arguments": [
+      "drug_name",
+      "indication",
+      "preclinical_data",
+      "target_population"
+    ],
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "drug_name": {
+          "type": "string",
+          "description": "Name of the drug candidate"
+        },
+        "indication": {
+          "type": "string",
+          "description": "Disease indication"
+        },
+        "preclinical_data": {
+          "type": "string",
+          "description": "Preclinical efficacy and safety data",
+          "default": ""
+        },
+        "target_population": {
+          "type": "string",
+          "description": "Target patient population",
+          "default": "General adult population"
+        }
+      },
+      "required": [
+        "drug_name",
+        "indication"
+      ]
+    },
+    "configs": {
+      "api_type": "CHATGPT",
+      "model_id": "gpt-4.1",
+      "temperature": 0.6,
+      "max_new_tokens": 5000,
+      "return_json": true
+    }
+  }
+]

tooluniverse/data/europe_pmc_tools.json

@@ -17,8 +17,7 @@
         }
       },
       "required": [
-        "query",
-        "limit"
+        "query"
       ]
     },
     "return_schema": {

tooluniverse/data/genomics_tools.json

@@ -0,0 +1,174 @@
+[
+  {
+    "type": "EnsemblTool",
+    "name": "Ensembl_lookup_gene_by_symbol",
+    "description": "Lookup Ensembl gene by species and gene symbol, returning core metadata and coordinates (uses /xrefs/symbol then /lookup/id?expand=1).",
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "species": {"type": "string", "description": "Species name (e.g., 'homo_sapiens').", "default": "homo_sapiens"},
+        "symbol": {"type": "string", "description": "Gene symbol (e.g., BRCA1)."}
+      },
+      "required": ["symbol"]
+    },
+    "return_schema": {
+      "type": "object",
+      "description": "Gene metadata including Ensembl ID, name, location, and transcripts count.",
+      "properties": {
+        "id": {"type": "string"},
+        "symbol": {"type": "string"},
+        "display_name": {"type": "string"},
+        "species": {"type": "string"},
+        "seq_region_name": {"type": "string"},
+        "start": {"type": "integer"},
+        "end": {"type": "integer"},
+        "strand": {"type": "integer"},
+        "biotype": {"type": "string"},
+        "transcript_count": {"type": "integer"}
+      }
+    }
+  },
+  {
+    "type": "ClinVarTool",
+    "name": "ClinVar_search_variants",
+    "description": "Search ClinVar via NCBI E-utilities (esearch→esummary) and return concise variant records for a query term.",
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "query": {"type": "string", "description": "ClinVar search term (e.g., BRCA1)."},
+        "retmax": {"type": "integer", "description": "Max records.", "default": 5}
+      },
+      "required": ["query"]
+    },
+    "return_schema": {
+      "type": "array",
+      "items": {
+        "type": "object",
+        "properties": {
+          "uid": {"type": "string"},
+          "accession": {"type": "string"},
+          "title": {"type": "string"},
+          "gene": {"type": "string"},
+          "chr": {"type": "string"},
+          "start": {"type": "integer"},
+          "stop": {"type": "integer"},
+          "spdi": {"type": "string"},
+          "clinical_significance": {"type": "string"}
+        }
+      }
+    }
+  },
+  {
+    "type": "DbSnpTool",
+    "name": "dbSNP_get_variant_by_rsid",
+    "description": "Fetch dbSNP variant by rsID using NCBI Variation Services (refsnp endpoint).",
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "rsid": {"type": "string", "description": "rsID without 'rs' prefix or with (e.g., rs699 or 699)."}
+      },
+      "required": ["rsid"]
+    },
+    "return_schema": {
+      "type": "object",
+      "properties": {
+        "refsnp_id": {"type": "string"},
+        "chrom": {"type": "string"},
+        "pos": {"type": "integer"},
+        "alleles": {"type": "array", "items": {"type": "string"}},
+        "hgvs": {"type": "array", "items": {"type": "string"}}
+      }
+    }
+  },
+  {
+    "type": "UCSCTool",
+    "name": "UCSC_get_genes_by_region",
+    "description": "Query UCSC Genome Browser track API for knownGene features in a genomic window.",
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "genome": {"type": "string", "description": "Genome assembly (e.g., hg38).", "default": "hg38"},
+        "chrom": {"type": "string", "description": "Chromosome (e.g., chr17)."},
+        "start": {"type": "integer", "description": "Start position (0-based)."},
+        "end": {"type": "integer", "description": "End position."},
+        "track": {"type": "string", "description": "Track name.", "default": "knownGene"}
+      },
+      "required": ["chrom", "start", "end"]
+    },
+    "return_schema": {
+      "type": "object",
+      "properties": {
+        "itemsReturned": {"type": "integer"},
+        "features": {
+          "type": "array",
+          "items": {
+            "type": "object",
+            "properties": {
+              "name": {"type": "string"},
+              "geneName": {"type": "string"},
+              "chrom": {"type": "string"},
+              "chromStart": {"type": "integer"},
+              "chromEnd": {"type": "integer"},
+              "strand": {"type": "string"}
+            }
+          }
+        }
+      }
+    }
+  },
+  {
+    "type": "GnomadTool",
+    "name": "gnomAD_query_variant",
+    "description": "Query gnomAD GraphQL for a variant in a dataset (returns ID and genome allele counts/frequency).",
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "variant_id": {"type": "string", "description": "Variant ID like '1-230710048-A-G'."},
+        "dataset": {"type": "string", "description": "Dataset ID (e.g., gnomad_r4).", "default": "gnomad_r4"}
+      },
+      "required": ["variant_id"]
+    },
+    "return_schema": {
+      "type": "object",
+      "properties": {
+        "variantId": {"type": "string"},
+        "genome": {
+          "type": "object",
+          "properties": {"ac": {"type": "integer"}, "an": {"type": "integer"}, "af": {"type": "number"}}
+        }
+      }
+    }
+  },
+  {
+    "type": "GWASGeneSearch",
+    "name": "GWAS_search_associations_by_gene",
+    "description": "Search GWAS Catalog associations by gene name (returns strongest risk allele and p-value fields).",
+    "parameter": {
+      "type": "object",
+      "properties": {
+        "gene_name": {"type": "string", "description": "Gene symbol (e.g., BRCA1)."},
+        "size": {"type": "integer", "description": "Max associations to return.", "default": 5}
+      },
+      "required": ["gene_name"]
+    },
+    "return_schema": {
+      "type": "object",
+      "properties": {
+        "gene_name": {"type": "string"},
+        "association_count": {"type": "integer"},
+        "associations": {
+          "type": "array",
+          "items": {
+            "type": "object",
+            "properties": {
+              "study": {"type": "string"},
+              "strongestRiskAllele": {"type": "string"},
+              "pvalue": {"type": "number"}
+            }
+          }
+        },
+        "total_found": {"type": "integer"}
+      }
+    }
+  }
+]