Claidex MCP
Biomedical failure intelligence platform for terminated trials and translational analysis.
使うべきか
品質と安全性
検出事項（1）
- LOWTool 'claidex_research_prompt' description lacks action verbclaidex_research_prompt 内
ツール定義とプロトコルへの準拠に関する自動分析に基づいています。
コンテキストコスト
これは、サーバーのツールがモデルのコンテキストに読み込まれるたびに消費されるおおよそのトークン数です。数が多いほど、ほかのタスクに使える注意が減ります。
インストール
ワンクリックインストール
これを `claude_desktop_config.json` ファイルに追加してください:
{
"mcpServers": {
"failure-intelligence": {
"command": "npx",
"args": [
"@claidex-mcp/claidex-mcp"
]
}
}
}実行可能なパッケージ
1.0.0stdioリモートエンドポイント
https://app.claidex.com/api/mcpstreamable-httpできること
ツール一覧
ツール（24）
⚪claidex_capabilities
Return the Claidex MCP feature map, configured storage/model providers, safety controls, resources, prompts, and tool counts.
入力スキーマ
{
"type": "object",
"properties": {},
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢claidex_claim_risk_matrix(query, limit)
Search claims and summarize MRS bands, trial phases, failure archetypes, and high-risk programs for rapid diligence.
入力スキーマ
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 1,
"description": "Drug, target, disease, sponsor, or failure archetype."
},
"limit": {
"default": 10,
"type": "integer",
"minimum": 1,
"maximum": 20
}
},
"required": [
"query",
"limit"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}⚪claidex_health_snapshot
Return Sentinel metrics, coverage, graph counts, recent automation runs, MRS bands, and preprint-watch summaries.
入力スキーマ
{
"type": "object",
"properties": {},
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}⚪claidex_research_prompt(objective, target_gene, disease, risk_tolerance)
Compose a rigorous, reusable investigation prompt that tells an MCP client which Claidex tools and resources to use.
入力スキーマ
{
"type": "object",
"properties": {
"objective": {
"type": "string",
"minLength": 1,
"description": "The investigation objective to turn into a reusable MCP prompt."
},
"target_gene": {
"description": "Optional HGNC target symbol.",
"type": "string"
},
"disease": {
"description": "Optional disease area.",
"type": "string"
},
"risk_tolerance": {
"default": "medium",
"type": "string",
"enum": [
"low",
"medium",
"high"
]
}
},
"required": [
"objective",
"risk_tolerance"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}⚪claidex_tool_catalog(include_schemas, source)
Return the complete MCP tool catalog, optionally including full input schemas and filtering native versus agent-backed tools.
入力スキーマ
{
"type": "object",
"properties": {
"include_schemas": {
"default": false,
"description": "Include full JSON Schemas in the catalog.",
"type": "boolean"
},
"source": {
"default": "all",
"type": "string",
"enum": [
"all",
"claidex-agent",
"mcp-native"
]
}
},
"required": [
"include_schemas",
"source"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}⚪consult_specialist_model(task_type, prompt, context)
Ask a specialist Gateway model for code, database/tool-calling, statistics, literature, preprint, or mechanistic synthesis support. Use for hard subproblems, second opinions, code generation/review, or complex synthesis.
入力スキーマ
{
"type": "object",
"properties": {
"task_type": {
"type": "string",
"enum": [
"code",
"database",
"reasoning",
"statistics",
"research",
"literature",
"synthesis",
"preprint",
"general"
],
"description": "Specialist routing target."
},
"prompt": {
"type": "string",
"minLength": 1,
"maxLength": 6000,
"description": "Focused task for the specialist model."
},
"context": {
"description": "Relevant context, snippets, or data. Do not include secrets.",
"type": "string",
"maxLength": 10000
}
},
"required": [
"task_type",
"prompt"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢fetch(id)
Fetch a specific result returned by search. This compatibility tool supports OpenAI and ChatGPT remote MCP retrieval flows.
入力スキーマ
{
"type": "object",
"properties": {
"id": {
"type": "string",
"minLength": 1,
"description": "Result id returned by the search tool."
}
},
"required": [
"id"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢fetch_research_url(url, max_chars)
Fetch a public HTTPS research URL, JSON API response, package metadata endpoint, documentation page, or dataset preview. Blocks private/internal hosts and truncates large responses.
入力スキーマ
{
"type": "object",
"properties": {
"url": {
"type": "string",
"format": "uri",
"description": "Public HTTPS URL to fetch."
},
"max_chars": {
"default": 8000,
"type": "number",
"minimum": 500,
"maximum": 20000
}
},
"required": [
"url",
"max_chars"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟡get_claim_content(slug)
Retrieve the complete text content of a specific Claidex Claim by its slug. Use this when you need to read the full post-mortem analysis, including hypothesis, failure mechanism, and prevention analysis.
入力スキーマ
{
"type": "object",
"properties": {
"slug": {
"type": "string",
"minLength": 1,
"description": "The Claidex claim slug, e.g. exicorilant-mcrpc-gr-bypass-fourth-failure"
}
},
"required": [
"slug"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢query_chembl(drug_name)
Look up a drug or compound in ChEMBL. Returns mechanism of action, primary target, ChEMBL ID, max development phase, molecular type, and synonyms.
入力スキーマ
{
"type": "object",
"properties": {
"drug_name": {
"type": "string",
"minLength": 1,
"description": "Drug or compound name"
}
},
"required": [
"drug_name"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢query_clinicaltrials(query, status, phase, max_results)
Search ClinicalTrials.gov for clinical trials. Can search by drug name, condition, target, or NCT ID. Returns trial status, phase, sponsor, primary endpoints, and why_stopped if terminated.
入力スキーマ
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 1,
"description": "Search term: drug name, condition, NCT ID, or target"
},
"status": {
"default": "any",
"type": "string",
"enum": [
"RECRUITING",
"ACTIVE_NOT_RECRUITING",
"TERMINATED",
"WITHDRAWN",
"COMPLETED",
"any"
]
},
"phase": {
"default": "any",
"type": "string",
"enum": [
"PHASE1",
"PHASE2",
"PHASE3",
"PHASE4",
"any"
]
},
"max_results": {
"default": 5,
"type": "number",
"minimum": 1,
"maximum": 10
}
},
"required": [
"query",
"status",
"phase",
"max_results"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢query_failure_graph(target_gene)
Query the Claidex failure graph for a specific gene target. Returns the Mechanism Risk Score (MRS), failure counts by phase and archetype, and all related claim slugs.
入力スキーマ
{
"type": "object",
"properties": {
"target_gene": {
"type": "string",
"minLength": 1,
"description": "HGNC gene symbol, e.g. PIK3CA, KRAS, EGFR"
}
},
"required": [
"target_gene"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢query_openfda(drug_name, max_results)
Query the FDA FAERS database for adverse event signals for a drug. Returns top adverse event terms, serious case count, and death count.
入力スキーマ
{
"type": "object",
"properties": {
"drug_name": {
"type": "string",
"minLength": 1,
"description": "Drug name as it appears in FAERS"
},
"max_results": {
"default": 10,
"type": "number",
"minimum": 1,
"maximum": 20
}
},
"required": [
"drug_name",
"max_results"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢query_opentargets(target_gene, disease_term)
Query Open Targets for target-disease association evidence. Returns genetic evidence score, clinical evidence, known drugs, tractability, and safety liabilities for a gene-disease pair.
入力スキーマ
{
"type": "object",
"properties": {
"target_gene": {
"type": "string",
"minLength": 1,
"description": "HGNC gene symbol"
},
"disease_term": {
"type": "string",
"minLength": 1,
"description": "Disease name or EFO term"
}
},
"required": [
"target_gene",
"disease_term"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢query_package_registry(ecosystem, package_name)
Look up package metadata from npm or PyPI before recommending code, analysis packages, API clients, or dataset tooling.
入力スキーマ
{
"type": "object",
"properties": {
"ecosystem": {
"type": "string",
"enum": [
"npm",
"pypi"
],
"description": "Package registry to query."
},
"package_name": {
"type": "string",
"minLength": 1,
"maxLength": 120,
"description": "Package name."
}
},
"required": [
"ecosystem",
"package_name"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢rank_documents_by_embedding(query, documents, top_k)
Embed a query and candidate documents, then rank documents by cosine similarity. Use for semantic matching, retrieval checks, clustering triage, and lightweight RAG over user-provided passages.
入力スキーマ
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 1,
"maxLength": 4000
},
"documents": {
"minItems": 1,
"maxItems": 30,
"type": "array",
"items": {
"type": "string",
"minLength": 1,
"maxLength": 8000
}
},
"top_k": {
"default": 10,
"type": "number",
"minimum": 1,
"maximum": 20
}
},
"required": [
"query",
"documents",
"top_k"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢rerank_documents(query, documents, top_n)
Rerank candidate documents/passages against a query using a dedicated reranking model. Use after retrieval/search when exact relevance ordering matters.
入力スキーマ
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 1,
"maxLength": 4000
},
"documents": {
"minItems": 1,
"maxItems": 50,
"type": "array",
"items": {
"type": "string",
"minLength": 1,
"maxLength": 8000
}
},
"top_n": {
"default": 10,
"type": "number",
"minimum": 1,
"maximum": 20
}
},
"required": [
"query",
"documents",
"top_n"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}⚪run_computation(expression, variables)
Evaluate deterministic mathematical expressions with optional variables. Use for algebra, matrices, vectors, transformations, descriptive numerical checks, and reproducible calculations. Does not execute arbitrary JavaScript or install packages.
入力スキーマ
{
"type": "object",
"properties": {
"expression": {
"type": "string",
"minLength": 1,
"maxLength": 4000,
"description": "MathJS expression, e.g. mean(values), std(values), multiply(A, b), inv(A)."
},
"variables": {
"default": {},
"description": "Variables available to the expression. Values must be numbers, strings, vectors, or matrices.",
"type": "object",
"propertyNames": {
"type": "string"
},
"additionalProperties": {
"anyOf": [
{
"type": "number"
},
{
"type": "string"
},
{
"type": "array",
"items": {
"type": "number"
}
},
{
"type": "array",
"items": {
"type": "array",
"items": {
"type": "number"
}
}
}
]
}
}
},
"required": [
"expression",
"variables"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}⚪run_statistical_analysis(analysis_type, inputs, context)
Run a statistical analysis computation. Supports: Wilson confidence intervals, binomial proportions, phase transition success rates, power calculations, Bayesian posterior estimates, descriptive statistics, and two-proportion comparisons. Returns computed values, interpretation, and LaTeX formula.
入力スキーマ
{
"type": "object",
"properties": {
"analysis_type": {
"type": "string",
"enum": [
"wilson_ci",
"phase_transition_rate",
"power_calculation",
"bayesian_posterior",
"descriptive_stats",
"failure_rate_comparison"
]
},
"inputs": {
"type": "object",
"propertyNames": {
"type": "string"
},
"additionalProperties": {
"anyOf": [
{
"type": "number"
},
{
"type": "string"
},
{
"type": "array",
"items": {
"type": "number"
}
}
]
},
"description": "Named input parameters. wilson_ci: {successes, n, confidence}. phase_transition_rate: {phase, indication, n_succeeded, n_total}. power_calculation: {n, effect_size, alpha}. bayesian_posterior: {prior_success_rate, n_observed, n_success, prior_strength?}. descriptive_stats: {values: number[]}. failure_rate_comparison: {a_success, a_n, b_success, b_n}."
},
"context": {
"description": "Optional context about what this analysis is for",
"type": "string"
}
},
"required": [
"analysis_type",
"inputs"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢search(query, limit)
Search Claidex failure claims and preprint flags. This compatibility tool returns citation-ready results for OpenAI and ChatGPT remote MCP flows.
入力スキーマ
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 1,
"description": "Biomedical failure intelligence query."
},
"limit": {
"default": 8,
"type": "integer",
"minimum": 1,
"maximum": 10
}
},
"required": [
"query",
"limit"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟡search_claims(query, category, failure_archetype, limit)
Search the Claidex failure database for post-mortems matching a drug name, gene target, disease, failure archetype, or sponsor. Returns matching claims with title, slug, drug, target, disease, phase, failure archetype, Open Targets score, and MRS score.
入力スキーマ
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 1,
"description": "Search term: drug name, gene symbol, disease, or failure archetype"
},
"category": {
"default": "any",
"type": "string",
"enum": [
"Oncology",
"Cardiology",
"Neurology",
"Immunology",
"Infectious Disease",
"Metabolic",
"Rare Disease",
"Other",
"any"
]
},
"failure_archetype": {
"default": "any",
"type": "string",
"enum": [
"efficacy_failure",
"safety_signal",
"enrollment_collapse",
"sponsor_decision",
"strategic_reprioritization",
"translational_mismatch",
"biomarker_failure",
"any"
]
},
"limit": {
"default": 10,
"type": "number",
"minimum": 1,
"maximum": 20
}
},
"required": [
"query",
"category",
"failure_archetype",
"limit"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢search_preprint_flags(query, severity)
Search the Claidex Preprint Watch database for bioRxiv preprints flagged as contradicting the clinical failure record. Search by target gene, disease, or preprint title.
入力スキーマ
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 1,
"description": "Search term: gene symbol, disease, or preprint title fragment"
},
"severity": {
"default": "any",
"type": "string",
"enum": [
"STRONG",
"MODERATE",
"MILD",
"any"
]
}
},
"required": [
"query",
"severity"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢search_pubmed(query, max_results)
Search PubMed for biomedical literature. Returns title, authors, journal, year, DOI, abstract, and PMID for each result.
入力スキーマ
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 1,
"description": "PubMed search query"
},
"max_results": {
"default": 5,
"type": "number",
"minimum": 1,
"maximum": 10
}
},
"required": [
"query",
"max_results"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}🟢web_search(query, focus)
Search the web for recent news, investor commentary, analyst notes, press releases, and conference coverage about drugs, clinical trials, or biomedical mechanisms.
入力スキーマ
{
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 1,
"description": "Search query"
},
"focus": {
"default": "any",
"type": "string",
"enum": [
"news",
"academic",
"investor",
"any"
]
}
},
"required": [
"query",
"focus"
],
"$schema": "https://json-schema.org/draft/2020-12/schema",
"additionalProperties": false
}コミュニティ
エビデンス