Summary
Brings pharmaceutical-grade health economics workflows into Claude as MCP tools. You get literature search across 44 sources (PubMed, NICE, ICER), automated risk-of-bias assessment using RoB 2 and ROBINS-I frameworks, cost-effectiveness modeling with EQ-5D value sets, and HTA dossier generation for NICE, FDA, EMA, and EU JCA submissions. The pharmacovigilance classifier maps studies to EMA regulatory categories and GVP modules in under 200ms. Includes workflow orchestrators for MAIC pipelines and IRB review prep. Built for pharma, biotech, and CRO teams who need auditable HEOR outputs without standing up custom infrastructure. Ships with JSON examples for complex tools and an AI transparency disclosure system aligned with ISPOR ELEVATE-GenAI guidelines.
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →
Tools
Public tool metadata for what this MCP can expose to an agent.
31 tools
literature.searchSearch PubMed, ClinicalTrials.gov, bioRxiv/medRxiv, ChEMBL, FDA Orange Book, FDA Purple Book, enterprise sources (Embase, ScienceDirect, Cochrane, Citeline, Pharmapendium, Cortellis), HTA cost reference sources (CMS NADAC, PSSRU, NHS National Cost Collection, BNF, PBS Schedule...8 params
Search PubMed, ClinicalTrials.gov, bioRxiv/medRxiv, ChEMBL, FDA Orange Book, FDA Purple Book, enterprise sources (Embase, ScienceDirect, Cochrane, Citeline, Pharmapendium, Cortellis), HTA cost reference sources (CMS NADAC, PSSRU, NHS National Cost Collection, BNF, PBS Schedule...
Parameters* required
runsnumberNumber of search runs (1-5, default 1). Multiple runs deduplicate and rank by consistency. Use runs=3 for comprehensive, stable results.
querystringResearch question (e.g. 'semaglutide type 2 diabetes cost-effectiveness')
projectstringProject ID for knowledge base persistence. When set, results are saved to ~/.heor-agent/projects/{project}/raw/literature/
sourcesarrayData sources to query. Default: pubmed, clinicaltrials, biorxiv, chembl, wiley, embase. Use 'who_gho' and 'world_bank' for epidemiology and demographic data. Use 'oecd' for OECD health statistics (expenditure, hospital beds, physicians, life expectancy). Use 'ihme_gbd' for Global Burden of Disease estimates. Use 'orange_book' for FDA drug approvals. Use 'purple_book' for FDA biologics and biosimilars. Enterprise (require API key): 'cochrane' (COCHRANE_API_KEY), 'citeline' (CITELINE_API_KEY), 'pharmapendium' (PHARMAPENDIUM_API_KEY), 'cortellis' (CORTELLIS_API_KEY). HTA cost refs: 'cms_nadac', 'pssru', 'nhs_costs', 'bnf', 'pbs_schedule'. LATAM sources: 'datasus', 'conitec', 'anvisa', 'paho', 'iets', 'fonasa'. APAC sources: 'hitap'. HTA appraisal/precedent sources: 'nice_ta', 'cadth_reviews', 'icer_reports', 'pbac_psd', 'gba_decisions', 'has_tc', 'iqwig', 'aifa', 'tlv', 'inesss'. HEOR methodology sources: 'ispor', 'wiley' (Pharmacoeconomics, Health Economics, Value in Health — via CrossRef), 'ohe' (Office of Health Economics — value set analyses, HTA methodology), 'euroqol' (EuroQol Group — EQ-5D instrument, country value sets, crosswalks).
date_fromstringExclude results before this date (ISO format: YYYY-MM-DD)
max_resultsnumberMaximum results to return (default: 20, max: 100)
study_typesarrayFilter results by study design. Values: 'rct' (randomised controlled trials), 'meta_analysis', 'observational' (cohort/case-control/cross-sectional), 'review' (systematic reviews, narrative reviews).
output_formatstringOutput format. 'docx' requires hosted tier.one of
text · json · docxmodels.cost_effectivenessBuild a cost-utility analysis (ICER, QALY, PSA, sensitivity analysis) for a drug vs comparator. Follows ISPOR good practice guidelines and NICE reference case. Includes probabilistic sensitivity analysis (PSA), one-way sensitivity, and cost-effectiveness acceptability curve (C...18 params
Build a cost-utility analysis (ICER, QALY, PSA, sensitivity analysis) for a drug vs comparator. Follows ISPOR good practice guidelines and NICE reference case. Includes probabilistic sensitivity analysis (PSA), one-way sensitivity, and cost-effectiveness acceptability curve (C...
Parameters* required
projectstringProject ID for knowledge base persistence. When set, model run is saved to ~/.heor-agent/projects/{project}/raw/models/
run_psabooleanRun probabilistic sensitivity analysis (default: true)
run_owsabooleanRun one-way sensitivity analysis (default: true). Set to false to skip OWSA and speed up large PSA runs.
scenariosarrayOptional scenario analysis: array of named parameter overrides. Each scenario re-runs the model with the specified changes. Max 10 scenarios.
comparatorstringComparator (standard of care)
indicationstringDisease or condition
model_typestringModel type. Default: markov. Use 'partsa' for oncology — requires survival_inputs (the handler now hard-fails when partsa is set without it).one of
markov · partsa · decision_treecost_inputsobjectAnnual costs per patient in the base currency for the selected perspective.
perspectivestringEconomic perspective: 'nhs' (UK NHS, WTP £25-35K/QALY), 'us_payer' ($100-150K/QALY ICER standard), or 'societal' (broader costs incl. productivity).one of
nhs · us_payer · societalinterventionstringDrug or treatment name
time_horizonstringModelling horizon: 'lifetime', '5yr', '10yr', or years as number
output_formatstringUse 'xlsx' for a structured Excel report with inputs, transition matrix, PSA iterations, and CEAC in separate tabs. The workbook is a REPORT — editing cells does not re-run the model. Re-run by calling the tool again with modified parameters.one of
text · json · docx · xlsxpsa_iterationsnumberPSA iterations (default: 1000, max: 10000)
summary_metricstringSummary metric for the ICER numerator. 'qaly' (default, NICE reference case), 'evlyg' (equal value life-years gained — CMS IRA-compatible; CMS prohibits QALYs in Medicare IRA drug price negotiations per §1194(e)(2)), or 'both' to report both side-by-side.one of
qaly · evlyg · bothutility_inputsobjectQALY weights for each health state (optional — defaults derived from efficacy if omitted).
clinical_inputsobjectClinical efficacy and safety parameters driving the Markov transitions.
mfn_sensitivityobjectOptional MFN price-sensitivity sweep. When supplied, the output includes an mfn_sensitivity block with ICER per price point and WTP-crossover prices. Design log #27.
survival_inputsobjectRequired when model_type='partsa'. Median survival inputs (months) for OS/PFS, optional comparator overrides, and parametric distribution choice. Codex P1 fix (2026-05-07).
hta.dossierStructure evidence into HTA body-specific submission format (NICE STA, EMA, FDA, IQWiG, HAS, EU JCA, or Global Value Dossier). Produces draft sections with gap analysis and auto-GRADE evidence quality tables. Accepts output from literature.search and models.cost_effectiveness....18 params
Structure evidence into HTA body-specific submission format (NICE STA, EMA, FDA, IQWiG, HAS, EU JCA, or Global Value Dossier). Produces draft sections with gap analysis and auto-GRADE evidence quality tables. Accepts output from literature.search and models.cost_effectiveness....
Parameters* required
picosarrayJCA-specific: list of PICOs from the scoping decision. Each PICO generates its own dossier section. If omitted, a default single PICO is generated.
projectstringProject ID for knowledge base persistence. When set, dossier draft is saved to ~/.heor-agent/projects/{project}/raw/dossiers/
hta_bodystringHTA body/dossier format. 'nice'=UK NICE STA; 'ema'=EMA CTD; 'fda'=FDA prescribing info; 'iqwig'=Germany AMNOG; 'has'=France transparency committee; 'jca'=EU Joint Clinical Assessment (Reg. 2021/2282); 'gvd'=Global Value Dossier (cross-market foundational document).one of
nice · ema · fda · iqwig · has · jcadrug_namestringGeneric or brand name of the drug/intervention.
indicationstringDisease or condition being treated (e.g., 'type 2 diabetes', 'non-small cell lung cancer').
mfn_contextobjectOptional MFN pricing context. When basket_prices is supplied and hta_body is one of {nice, ema, fda, iqwig, has, jca, gvd}, renders an MFN Exposure section with ceiling math and gap-to-US analysis. Design log #27.
rob_resultsvalueOutput from the evidence.risk_of_bias tool. When provided, the GRADE Risk of Bias domain uses the structured judgment (rob_judgment, downgrade, rationale) instead of a heuristic estimate.
model_resultsvalueJSON output from models.cost_effectiveness — used to populate the Economic Evidence Summary section.
output_formatstringOutput format. 'text' returns markdown; 'json' returns structured sections; 'docx' generates a Word document and saves to disk.one of
text · json · docxsubmission_typestringSubmission type. 'sta'/'mta' for NICE; 'initial'/'renewal'/'variation' for JCA; 'early_access' for accelerated pathways.one of
sta · mta · early_access · initial · renewal · variationevidence_summaryvalueClinical evidence input. Accepts: a text summary string, OR a JSON array of LiteratureResult objects from literature.search (use output_format='json'). When passed as array, auto-generates a GRADE evidence quality table.
pv_classificationobjectOptional: structured PV classification from the pv_classify tool. When provided, the dossier includes a Pharmacovigilance Plan section with GVP module, ENCePP template, submission obligations, and RMP implications. Pipe pv_classify output here.
severity_modifierobjectNICE PMG36 severity modifier inputs (replaced end-of-life modifier, April 2022). Provide absolute_qaly_shortfall (years) and/or proportional_qaly_shortfall (0-1). Modifier weight: <12 absolute AND <0.85 proportional → 1.0×; 12-18 or 0.85-0.95 → 1.2×; ≥18 or ≥0.95 → 1.7×.
unmet_need_summarystringOptional: 1-paragraph unmet need synthesis from the evidence.unmet_need tool. Pipe evidence.unmet_need result.unmet_need_summary here. Prepended to the Unmet Need section for NICE (Section B) and GVD (Section 4).
health_inequalitiesobjectHealth inequalities evidence per NICE PMG36 May 2025 modular update. Required by NICE for interventions affecting disadvantaged groups. intervention_impact: 'narrows'/'neutral'/'widens'/'unknown'.
regulatory_landscapearrayOptional: array of RegulatoryStatusResult objects from regulatory.status_check (via hta_workflow Phase 3.6). When provided and hta_body in {nice, jca, gvd}, renders a Regulatory Landscape section with a comparator × region × status table. Design log #26.
upgrading_per_outcomeobjectOptional: GRADE upgrading flags per outcome for observational evidence (Guyatt 2011). Keys are outcome names; values specify large_effect ('none'/'large'/'very_large'), dose_response (boolean), and plausible_confounding_toward_null (boolean). Ignored for RCT evidence. Capped at +2 steps.
heterogeneity_per_outcomeobjectOptional: I² and study count per outcome (from evidence_indirect tool). When provided, GRADE inconsistency is computed from I² instead of heuristic. Example: { 'overall survival': { i_squared_pct: 45, n_studies: 6 } }
knowledge.searchSearch a project's knowledge base (raw/ and wiki/) for text matches. Returns file paths with line numbers and snippets. Use this to find previously-retrieved literature, model runs, and compiled wiki content without re-querying external APIs.5 params
Search a project's knowledge base (raw/ and wiki/) for text matches. Returns file paths with line numbers and snippets. Use this to find previously-retrieved literature, model runs, and compiled wiki content without re-querying external APIs.
Parameters* required
pathsarrayWhich subtrees to search. Default: both.
querystringSearch query — multi-term searches match ANY term (OR)
projectstringProject ID (must exist)
max_resultsnumberMax matches to return (default 20, max 100)
case_sensitivebooleanCase-sensitive search (default false)
knowledge.readRead a file from a project's raw/ or wiki/ tree. Path is relative to project root. Only raw/ and wiki/ subtrees accessible.2 params
Read a file from a project's raw/ or wiki/ tree. Path is relative to project root. Only raw/ and wiki/ subtrees accessible.
Parameters* required
pathstringRelative path (e.g. 'wiki/trials/sustain-6.md' or 'raw/literature/pubmed_12345.md')
projectstringProject ID
knowledge.writeWrite a file to the project's wiki/ tree. Path MUST start with 'wiki/' and end with '.md'. Use this to compile/organize evidence from raw/ files into a structured knowledge base. Supports Obsidian-style [[wikilinks]].3 params
Write a file to the project's wiki/ tree. Path MUST start with 'wiki/' and end with '.md'. Use this to compile/organize evidence from raw/ files into a structured knowledge base. Supports Obsidian-style [[wikilinks]].
Parameters* required
pathstringRelative path starting with 'wiki/', ending with .md (e.g. 'wiki/trials/sustain-6.md')
contentstringMarkdown content. Can include YAML frontmatter and [[wikilinks]].
projectstringProject ID
project.createInitialize a new HEOR project workspace with directory skeleton and project.yaml metadata. Idempotent — returns existing project if already created. Required before using the `project` parameter in other tools. Enum values are case-insensitive — `"NICE"`, `"Nice"`, and `"nice"...5 params
Initialize a new HEOR project workspace with directory skeleton and project.yaml metadata. Idempotent — returns existing project if already created. Required before using the `project` parameter in other tools. Enum values are case-insensitive — `"NICE"`, `"Nice"`, and `"nice"...
Parameters* required
drugstringDrug or intervention name
notesstringFree-text project notes (optional)
indicationstringDisease/condition being treated
project_idstringShort identifier (alphanumeric + hyphens, e.g. 'semaglutide-t2d')
hta_targetsarrayHTA bodies to target (optional). Case-insensitive — "NICE"/"Nice"/"nice" are all accepted.
evidence.networkAnalyze literature search results to build an evidence network map. Extracts intervention-comparator pairs from titles and abstracts, constructs a treatment comparison network, and assesses NMA (network meta-analysis) feasibility. Pass the results array from a prior literature...2 params
Analyze literature search results to build an evidence network map. Extracts intervention-comparator pairs from titles and abstracts, constructs a treatment comparison network, and assesses NMA (network meta-analysis) feasibility. Pass the results array from a prior literature...
Parameters* required
querystringOriginal search query (optional, for context)
resultsarrayArray of LiteratureResult objects from a prior literature.search call (use output_format='json')
evidence.indirectCompute indirect treatment comparisons using the Bucher method (single common comparator) or frequentist network meta-analysis (full network). Requires user-supplied effect sizes (point estimates + 95% CI) from published trials. Supports MD, OR, RR, HR. Auto-selects method by...3 params
Compute indirect treatment comparisons using the Bucher method (single common comparator) or frequentist network meta-analysis (full network). Requires user-supplied effect sizes (point estimates + 95% CI) from published trials. Supports MD, OR, RR, HR. Auto-selects method by...
Parameters* required
methodstringauto (default): Bucher for simple paths, Frequentist NMA for 3+ edges. Or force a specific method.one of
auto · bucher · frequentist_nmatargetobjectOptional: specific comparison to compute. Omit to compute all possible pairwise comparisons.
comparisonsarrayDirect comparisons with effect sizes. Need at least 2 sharing a common comparator. Ask the user for: point estimate, 95% CI, outcome name, and effect measure (MD/OR/RR/HR) from each trial.
models.budget_impactEstimate the total budget impact of adopting a new intervention over 1-5 years. Follows ISPOR Budget Impact Analysis good practice guidelines (Mauskopf 2007, Sullivan 2014). Computes year-by-year net cost to payer, including market share uptake, treatment displacement, and pop...17 params
Estimate the total budget impact of adopting a new intervention over 1-5 years. Follows ISPOR Budget Impact Analysis good practice guidelines (Mauskopf 2007, Sullivan 2014). Computes year-by-year net cost to payer, including market share uptake, treatment displacement, and pop...
Parameters* required
projectstringProject ID for persistence
comparatorstringCurrent standard of care
indicationstringDisease or condition
perspectivestringPayer perspective: 'nhs' (UK NHS), 'us_payer' (US commercial/Medicare), or 'societal' (broader costs incl. productivity). Determines currency and cost categories.one of
nhs · us_payer · societaldisplacementarrayExisting treatments displaced by intervention (share of patients switching, cost saved)
interventionstringNew drug or treatment
market_shareobjectExpected market share of intervention by year (0-1). Missing years extrapolate from last defined.
output_formatstringUse 'xlsx' for a structured Excel report with all inputs, year-by-year calculations, and audit trail in separate tabs. The workbook is a REPORT — editing cells does not re-run the model. Re-run by calling the tool again with modified parameters.one of
text · json · docx · xlsxae_cost_annualnumberAnnual adverse event cost for intervention. Default 0.
drug_cost_annualnumberAnnual drug acquisition cost for intervention
admin_cost_annualnumberAnnual administration cost (applies to both arms). Default 0.
time_horizon_yearsnumberBudget horizon in years (1-10, default 5)
eligible_populationnumberNumber of eligible patients in Year 1
comparator_cost_annualnumberAnnual drug cost for comparator
monitoring_cost_annualnumberAnnual monitoring cost for intervention. Default 0.
population_growth_ratenumberAnnual population growth rate (e.g., 0.02 for 2%). Default 0.
comparator_ae_cost_annualnumberAnnual adverse event cost for comparator. Default 0.
evidence.population_adjusted⚠️ EXPERIMENTAL / orientation-only. Approximate population-adjusted indirect comparison using summary-level statistics (mean, SD per covariate). True MAIC/STC per NICE DSU TSD 18 requires individual patient data (IPD) for one trial. This tool inflates the SE of a Bucher indire...7 params
⚠️ EXPERIMENTAL / orientation-only. Approximate population-adjusted indirect comparison using summary-level statistics (mean, SD per covariate). True MAIC/STC per NICE DSU TSD 18 requires individual patient data (IPD) for one trial. This tool inflates the SE of a Bucher indire...
Parameters* required
methodstringauto (default): MAIC when >=2 modifiers and N>=50, else STCone of
auto · maic · stcprojectstringProject ID for persistence
index_trialobjectTrial with data to be reweighted (the trial for which you want to adjust)
outcome_namestringName of the outcome being compared (e.g., 'HbA1c change')
target_trialobjectTrial whose population is the matching target
output_formatstringone of
text · jsoneffect_modifiersarrayNames of covariates that are effect modifiers (must appear in both trials' covariates)
evidence.survivalFit parametric survival distributions (Exponential, Weibull, Log-logistic, Log-normal, Gompertz) to either patient-level event-time data (preferred — true right-censored MLE per Collett 2015 / NICE DSU TSD 14) OR Kaplan-Meier step-summary data (legacy approximation, used when...6 params
Fit parametric survival distributions (Exponential, Weibull, Log-logistic, Log-normal, Gompertz) to either patient-level event-time data (preferred — true right-censored MLE per Collett 2015 / NICE DSU TSD 14) OR Kaplan-Meier step-summary data (legacy approximation, used when...
Parameters* required
km_dataarrayKaplan-Meier step-summary data points (LEGACY approximation; use event_data when patient-level data is available). At least 3 required. Extract from published KM curves or trial reports.
projectstringProject ID for persistence
endpointstringEndpoint name (e.g., 'OS', 'PFS', 'DFS'). Default: 'OS'.
time_unitstringTime unit for KM data (default: months)one of
months · yearsevent_dataarrayPatient-level event-time rows (PREFERRED — true MLE per NICE DSU TSD 14). Each row: { time, event } where event=1 for an observed event, event=0 for right-censoring. At least 5 rows required.
output_formatstringone of
text · jsonliterature.screenScreen literature search results using PICO criteria. Scores each abstract by relevance to the research question, classifies study design, and returns a ranked shortlist with inclusion/exclusion decisions and reasons. Pass the results array from a prior literature.search call...7 params
Screen literature search results using PICO criteria. Scores each abstract by relevance to the research question, classifies study design, and returns a ranked shortlist with inclusion/exclusion decisions and reasons. Pass the results array from a prior literature.search call...
Parameters* required
projectstringProject ID for persistence
resultsarrayArray of LiteratureResult objects from a prior literature.search call (use output_format='json')
criteriaobjectPICO inclusion criteria
min_yearnumberExclude studies published before this year
output_formatstringone of
text · jsoninclude_thresholdnumberRelevance score threshold for inclusion (0-1, default 0.3). Lower = more inclusive.
exclude_study_typesarrayAdditional study types to exclude (e.g., ['case_report', 'narrative_review']). Editorials/commentaries are always excluded.
evidence.risk_of_biasAssess risk of bias for a set of studies using the appropriate Cochrane instrument: RoB 2 (RCTs), ROBINS-I (observational studies), or AMSTAR-2 (systematic reviews/meta-analyses). Instrument is auto-detected from study_type or can be specified. Judgments are inferred from abst...4 params
Assess risk of bias for a set of studies using the appropriate Cochrane instrument: RoB 2 (RCTs), ROBINS-I (observational studies), or AMSTAR-2 (systematic reviews/meta-analyses). Instrument is auto-detected from study_type or can be specified. Judgments are inferred from abst...
Parameters* required
studiesarrayArray of study objects from screen_abstracts or literature_search (output_format='json')
outcomesarrayOutcomes of interest for contextual domain assessment (optional)
instrumentstringAssessment instrument. auto (default) detects from study_type. rob2 for RCTs, robins_i for observational, amstar2 for systematic reviews.one of
auto · rob2 · robins_i · amstar2output_formatstringone of
text · jsonutils.validate_linksValidate URLs by making HEAD requests and checking HTTP status codes. Returns categorization: working (200), browser_only (403 from bot-blocking sites that work in browsers), broken (404/410), or timeout/error. ALWAYS use this before presenting reference links to users — broke...2 params
Validate URLs by making HEAD requests and checking HTTP status codes. Returns categorization: working (200), browser_only (403 from bot-blocking sites that work in browsers), broken (404/410), or timeout/error. ALWAYS use this before presenting reference links to users — broke...
Parameters* required
urlsarrayList of URLs to validate (max 50)
timeout_msnumberTimeout per URL in ms (default 10000, max 30000)
hta.utilityLook up EQ-5D value set characteristics (UK 3L, England 5L, new UK 5L 2026, NICE DSU mapping) or estimate the ICER/QALY impact of the new UK EQ-5D-5L value set for a given indication type. Cites Biz, Hernández Alava, Wailoo (2026) Value in Health. Use when user asks about NICE...5 params
Look up EQ-5D value set characteristics (UK 3L, England 5L, new UK 5L 2026, NICE DSU mapping) or estimate the ICER/QALY impact of the new UK EQ-5D-5L value set for a given indication type. Cites Biz, Hernández Alava, Wailoo (2026) Value in Health. Use when user asks about NICE...
Parameters* required
actionstring'lookup' returns a single value set; 'compare' returns all four side-by-side; 'estimate_impact' returns ICER/QALY change estimates for an indication type.one of
lookup · compare · estimate_impactbase_icernumberOptional: current ICER to project forward under new UK 5L.
value_setstringValue set id (required for 'lookup'). 'uk_5l_new' is the 2026 one under NICE consultation.one of
uk_3l · england_5l · uk_5l_new · dsu_mappingindication_typestringIndication category (required for 'estimate_impact'). 'non_cancer_qol_only' = chronic QoL-only conditions (migraine, UC, atopic dermatitis, HS, plaque psoriasis) — sees the biggest ICER increase.one of
cancer_life_extending · non_cancer_life_extending · non_cancer_qol_onlybase_incremental_qalynumberOptional: current incremental QALY gain to project forward under new UK 5L.
evidence.itcAssess the feasibility of an indirect treatment comparison (ITC) by walking through the three core assumptions (exchangeability, homogeneity, consistency) and recommending an appropriate method: direct comparison, Bucher, full NMA, anchored MAIC/STC, unanchored MAIC/STC, ML-NM...9 params
Assess the feasibility of an indirect treatment comparison (ITC) by walking through the three core assumptions (exchangeability, homogeneity, consistency) and recommending an appropriate method: direct comparison, Bucher, full NMA, anchored MAIC/STC, unanchored MAIC/STC, ML-NM...
Parameters* required
outcome_typestringPrimary outcome type — guides estimator recommendations.one of
binary · continuous · time_to_eventh2h_availablebooleanTrue if at least one head-to-head RCT compares the treatments directly. Default false.
connected_networkbooleanTrue if there is a connected evidence network linking the treatments of interest via at least one common comparator.
heterogeneity_i2_pctnumberOptional: I² statistic (%) across studies of the same comparison. If absent, homogeneity is assessed qualitatively.
subgroup_data_availablebooleanTrue if subgroup data are available in comparator trials for adjustment. Default false.
n_studies_per_comparisonnumberOptional: minimum number of studies per pairwise comparison. Informs choice between Bucher (k=1) and NMA (k≥2).
effect_modifier_imbalancestringSeverity of imbalance in identified effect modifiers across trial populations. Default 'unknown'.one of
none · minor · major · unknowneffect_modifiers_identifiedbooleanTrue if effect modifiers have been identified through clinical input or literature review. Default false.
ipd_available_for_interventionbooleanTrue if individual patient data (IPD) are available from the sponsor's trial. Default false.
examplesGet a pre-filled, copy-runnable JSON input for any of the heavy-schema tools (cost_effectiveness_model, budget_impact_model, survival_fitting, population_adjusted_comparison, evidence_indirect). Use when you want to demo a tool but don't want to invent inputs from scratch — th...1 params
Get a pre-filled, copy-runnable JSON input for any of the heavy-schema tools (cost_effectiveness_model, budget_impact_model, survival_fitting, population_adjusted_comparison, evidence_indirect). Use when you want to demo a tool but don't want to invent inputs from scratch — th...
Parameters* required
toolstringWhich tool to get an example for. Omit to list all available examples.one of
cost_effectiveness_model · budget_impact_model · survival_fitting · population_adjusted_comparison · evidence_indirectworkflow.maicRun the canonical MAIC discovery+screening pipeline in one call: ITC feasibility + parallel literature_search (broad + per-trial) + PICO screening + risk_of_bias + evidence_network. Stops short of running MAIC/Bucher itself (those require IPD or trial-level effect estimates)....10 params
Run the canonical MAIC discovery+screening pipeline in one call: ITC feasibility + parallel literature_search (broad + per-trial) + PICO screening + risk_of_bias + evidence_network. Stops short of running MAIC/Bucher itself (those require IPD or trial-level effect estimates)....
Parameters* required
picoobjectOptional PICO criteria for screening (defaults derived).
comparatorstringComparator drug name.
indicationstringDisease/condition.
interventionstringDrug/intervention name.
outcome_typestringone of
binary · continuous · time_to_eventruns_per_searchnumberdefault: 2
effect_modifiersarrayOptional effect modifiers known from clinical input.
trials_comparatorarrayOptional comparator-side trial names (e.g., INSPIRE, COMMAND).
trials_interventionarrayOptional trial names (e.g., QUASAR, ASTRO) to search in parallel.
max_results_per_searchnumberdefault: 50
pv.classifyClassify a planned study into its EMA pharmacovigilance regulatory category (PASS imposed/voluntary, PAES, RMP Annex 4, DUS, active surveillance registry, pregnancy registry, spontaneous reporting, ICH E2E plan). Returns the matching GVP module + ENCePP study-category label (n...10 params
Classify a planned study into its EMA pharmacovigilance regulatory category (PASS imposed/voluntary, PAES, RMP Annex 4, DUS, active surveillance registry, pregnancy registry, spontaneous reporting, ICH E2E plan). Returns the matching GVP module + ENCePP study-category label (n...
Parameters* required
drugstringDrug name.
indicationstringDisease/condition.
study_designstringone of
rct · single_arm · prospective_cohort · retrospective_cohort · case_control · registryjurisdictionsarraymulti_countrybooleandefault: false
primary_objectivestringone of
safety · efficacy · effectiveness · drug_utilization · natural_history · risk_minimisation_evaluationregulatory_contextstringone of
pre_authorisation · post_authorisation · conditional_approval · accelerated_approval · rmp_commitmentimposed_by_authoritybooleandefault: false
population_includes_pregnantbooleandefault: false
population_includes_paediatricbooleandefault: false
jca.pico_scopeProduce the canonical EU Joint Clinical Assessment (JCA) PICO matrix for a drug-indication pair. Returns a consolidated PICO list (per JCA process under Reg. 2021/2282) plus country-specific comparator universes, outcome instrument preferences, population subgroup focus, and a...10 params
Produce the canonical EU Joint Clinical Assessment (JCA) PICO matrix for a drug-indication pair. Returns a consolidated PICO list (per JCA process under Reg. 2021/2282) plus country-specific comparator universes, outcome instrument preferences, population subgroup focus, and a...
Parameters* required
drugstringDrug name.
is_orphanbooleanOptional. Orphan-designated medicinal product. Affects JCA scope eligibility (orphans enter scope from 13 January 2028 vs 13 January 2030 for general medicinal products).
drug_classstringone of
monoclonal_antibody · small_molecule · atmp_cell · atmp_gene · atmp_tissue · biosimilarindicationstringDisease/condition.
jurisdictionsarrayline_of_therapystringone of
first_line · second_line · third_line_plus · anydefault: anybiomarker_statusstringOptional biomarker status (e.g., 'EGFR T790M positive', 'PD-L1 TPS ≥50%').
regulatory_contextstringone of
pre_authorisation · post_authorisation · conditional_approvaldefault: post_authorisationmechanism_of_actionstringforce_proceed_out_of_scopebooleanOverride the JCA scope eligibility check. Default false. Set true to produce a JCA-style matrix anyway when the indication is not yet in JCA scope (e.g., for protocol-design or anticipatory market-access work). Output will carry an explicit out-of-scope warning either way.default: false
pv.signal_workflowCompute disproportionality statistics (PRR, ROR, IC/BCPNN, MGPS/EBGM) on user-supplied drug-AE case counts and decide a signal verdict per EMA GVP Module IX rev 2. Returns the verdict (no/strengthening/confirmed/previously known/refuted), workflow recommendations, and canonica...9 params
Compute disproportionality statistics (PRR, ROR, IC/BCPNN, MGPS/EBGM) on user-supplied drug-AE case counts and decide a signal verdict per EMA GVP Module IX rev 2. Returns the verdict (no/strengthening/confirmed/previously known/refuted), workflow recommendations, and canonica...
Parameters* required
drugstringindicationstringcase_countsobjectdata_sourcestringone of
eudravigilance · faers · vigibase_who · national_db · spontaneous_internaldefault: eudravigilanceoutcome_seriousbooleandefault: false
pregnancy_exposurebooleandefault: false
prior_known_signalsarrayreporting_period_monthsnumberdefault: 12
rmp_has_pregnancy_concernbooleandefault: false
hta.workflowEnd-to-end HTA submission orchestrator. One call runs literature_search (with PRISMA-style stability via runs=N) → screen_abstracts (PICO filter) → risk_of_bias (auto RoB 2/ROBINS-I/AMSTAR-2) → cost_effectiveness_model (Markov + 1k PSA, defaults from indication) → hta_dossier...16 params
End-to-end HTA submission orchestrator. One call runs literature_search (with PRISMA-style stability via runs=N) → screen_abstracts (PICO filter) → risk_of_bias (auto RoB 2/ROBINS-I/AMSTAR-2) → cost_effectiveness_model (Markov + 1k PSA, defaults from indication) → hta_dossier...
Parameters* required
drugstringDrug or intervention name.
picoobjectOptional PICO criteria for the screening phase (defaults derived from drug + indication).
sourcesarrayhta_bodystringTarget HTA body for the dossier draft.one of
nice · ema · fda · iqwig · has · jcadefault: nicece_inputsobjectOptional cost-effectiveness model inputs. Sensible defaults applied when omitted (efficacy_delta=0.25; drug_cost_annual=1000; SoC=0).
is_orphanbooleanOrphan-designated medicinal product. Affects JCA scope eligibility (orphans enter JCA scope 2028 vs 2030 for general medicinal products).default: false
drug_classstringone of
monoclonal_antibody · small_molecule · atmp_cell · atmp_gene · atmp_tissue · biosimilardefault: small_moleculeindicationstringDisease or condition (free text).
perspectivestringEconomic perspective for the cost-effectiveness model phase.one of
nhs · us_payer · societaldefault: nhsjurisdictionsarrayskip_ce_modelbooleanSkip the cost-effectiveness model phase (e.g., for clinical-only dossiers).default: false
literature_runsnumberNumber of dedup-stability passes for the literature_search phase. 3 is the project default for HTA-grade reproducibility.default: 3
submission_typestringone of
sta · mta · early_access · initial · renewal · variationdefault: staunmet_need_inputsobjectOptional structured inputs for Phase 3.5 (evidence.unmet_need). Only consumed when hta_body='gvd'. When supplied, the orchestrator runs evidence.unmet_need and pipes the resulting unmet_need_summary into Phase 5 hta_dossier (pre-filling GVD Section 4 / NICE Section B). Design log #23.
auto_check_regulatorybooleanWhen true (default), Phase 3.6 fans out regulatory.status_check across comparators from pico.comparator and unmet_need_inputs.treatment_landscape.current_soc. Results piped into hta_dossier as regulatory_landscape (renders 'Regulatory Landscape' section for nice/jca/gvd/amcp). Degrades gracefully on API errors. Set false to skip. Design log #26.default: true
max_literature_resultsnumberdefault: 30
irb.reviewClassify a planned study under 45 CFR 46 (US Common Rule) + EU CTR 536/2014 to produce an IRB / Ethics Committee submission scaffold. Returns: review tier (exempt §46.104 cat 1-8 / expedited §46.110 cat 1-7 / full-board §46.108), EU CTR review path with timeline, vulnerable-po...24 params
Classify a planned study under 45 CFR 46 (US Common Rule) + EU CTR 536/2014 to produce an IRB / Ethics Committee submission scaffold. Returns: review tier (exempt §46.104 cat 1-8 / expedited §46.110 cat 1-7 / full-board §46.108), EU CTR review path with timeline, vulnerable-po...
Parameters* required
indicationstringmulti_sitebooleandefault: false
risk_levelstringone of
minimal · greater_than_minimal · unknowndefault: unknowntaste_testbooleanSet true for taste / food-quality / consumer-acceptance evaluation (45 CFR 46.104(d)(6)).
interventionstringstudy_designstringone of
interventional · non_interventional_prospective · retrospective_chart_review · registry · secondary_data_analysis · specimen_repositorydata_handlingstringone of
fully_identifiable · pseudonymized · anonymized_safe_harbor · anonymized_expert_determination · aggregate_onlyjurisdictionsarraymarketed_drugbooleanSet true when the study uses a marketed drug per labeling (drives expedited cat 1, post-marketing PSUR framework).
funding_sourcestringone of
industry · nih · other_government · foundation · academic · nonepv_classificationobjectOptional structured output from pv_classify. Only primary_category is read; PASS_imposed triggers CTR Annex III SAE timelines.
benign_behaviouralbooleanSet true for benign behavioural interventions in adults (45 CFR 46.104(d)(3)).
exempt_category_hintstringHint: flag this as 45 CFR 46.104(d)(1) educational practices research.one of
educationalrecording_collectionbooleanSet true for voice / digital / image recording collection (45 CFR 46.110 cat 6).
federal_demonstrationbooleanSet true for federally supported research / demonstration project (45 CFR 46.104(d)(5)).
noninvasive_procedurebooleanSet true for noninvasive data collection (ECG, EEG, blood pressure, ultrasound) (45 CFR 46.110 cat 4).
broad_consent_obtainedbooleanSet true when broad consent has been obtained for secondary research / specimen storage (45 CFR 46.104(d)(7-8)).
noninvasive_collectionbooleanSet true for prospective biospecimen collection by noninvasive means (45 CFR 46.110 cat 3).
blood_draw_within_limitsbooleanSet true for blood collection within OHRP volume limits (45 CFR 46.110 cat 2).
expedited_category_claimarrayExpedited review categories per 45 CFR 46.110 (1-7) the investigator believes apply. Tool will validate; if it disagrees, both are surfaced as advisory_warnings.
population_includes_pregnantbooleandefault: false
population_includes_pediatricbooleandefault: false
population_includes_prisonersbooleandefault: false
population_includes_decisionally_impairedbooleandefault: false
icf.readability_checkScore the readability of an Informed Consent Form (ICF) text. Returns Flesch-Kincaid Grade Level, Flesch Reading Ease, Gunning Fog Index, SMOG Grade, plus per-sentence breakdown identifying the worst offenders, medical-jargon detection with plain-language alternatives, and a p...3 params
Score the readability of an Informed Consent Form (ICF) text. Returns Flesch-Kincaid Grade Level, Flesch Reading Ease, Gunning Fog Index, SMOG Grade, plus per-sentence breakdown identifying the worst offenders, medical-jargon detection with plain-language alternatives, and a p...
Parameters* required
icf_textstringThe ICF body text. Plain text or markdown both work.
jargon_checkbooleanWhether to run the medical-jargon dictionary lookup. Default true.default: true
target_grade_levelintegerTarget US grade level for FKGL pass verdict. FDA/NIH typically 8; some IRBs use 6 (stricter).default: 8
evidence.clinical_scaleScore neurology & cognitive outcome scales (UMSARS/UPDRS/MDS-UPDRS/ADAS-Cog/MoCA/MMSE). Returns total + subscale scores, MCID-based responder classification, and trajectory comparison vs NNIPPS/PPMI/ADNI reference cohorts. Integrates with jca_pico_scope for MSA (neurology_msa,...5 params
Score neurology & cognitive outcome scales (UMSARS/UPDRS/MDS-UPDRS/ADAS-Cog/MoCA/MMSE). Returns total + subscale scores, MCID-based responder classification, and trajectory comparison vs NNIPPS/PPMI/ADNI reference cohorts. Integrates with jca_pico_scope for MSA (neurology_msa,...
Parameters* required
itemsarrayPer-item scores. Supply all items for complete scoring, or a subset for partial scoring with a note.
scalestringClinical outcome scale: umsars | updrs | mds_updrs | adas_cog | moca | mmseone of
umsars · updrs · mds_updrs · adas_cog · moca · mmsebaseline_itemsarrayOptional: baseline scores for change-from-baseline and responder analysis.
compare_cohortstringNatural-history reference cohort for trajectory comparison. nnipps=MSA, ppmi=PD, adni=AD.one of
nnipps · ppmi · adni · nonetime_point_monthsnumberTime point in months from baseline (used for trajectory comparison).
evidence.unmet_needGenerate a structured unmet need section for HTA dossiers (NICE STA, EMA, FDA, IQWiG, HAS, JCA, GVD, AMCP). Consume-only: first retrieve evidence with literature_search, then pass only cited facts across 4 HEOR dimensions — disease burden, treatment landscape, QoL impact, econ...9 params
Generate a structured unmet need section for HTA dossiers (NICE STA, EMA, FDA, IQWiG, HAS, JCA, GVD, AMCP). Consume-only: first retrieve evidence with literature_search, then pass only cited facts across 4 HEOR dimensions — disease burden, treatment landscape, QoL impact, econ...
Parameters* required
drugstringDrug or intervention name
indicationstringDisease/condition being treated
qol_impactobjectPatient-reported and health-related quality of life impact
jurisdictionsarrayTarget market(s) for this unmet need assessment
disease_burdenobjectEpidemiology data for the indication
economic_burdenobjectDirect and indirect economic burden of the disease
literature_evidencearrayOptional: pass literature_search results here to include as supporting evidence
treatment_landscapeobjectCurrent standard of care and its limitations. Use only evidence retrieved by literature_search or other audited sources. Regulatory/approval claims require current label/regulatory citation support; if not confirmed, leave out rather than infer.
auto_check_regulatorybooleanWhen true (default), automatically fans out regulatory.status_check for each drug in treatment_landscape.current_soc × inferred regions. Injects verbatim label quotes inline with auto-numbered citations. Set false to skip (pre-v1.10.1 behaviour). Design log #26.default: true
regulatory.status_checkLook up current regulatory approval status for a drug from primary sources (OpenFDA for US, EMA EPI for EU). Returns approved indications, label text verbatim, age/weight/sex constraints, black-box warnings, REMS status, contraindications, source URLs, and fetch timestamp. Ref...5 params
Look up current regulatory approval status for a drug from primary sources (OpenFDA for US, EMA EPI for EU). Returns approved indications, label text verbatim, age/weight/sex constraints, black-box warnings, REMS status, contraindications, source URLs, and fetch timestamp. Ref...
Parameters* required
drugstringDrug name: INN (fremanezumab), INN+suffix (fremanezumab-vfrm), or brand (Ajovy, case-insensitive).
regionstringRegulatory region. 'uk' deferred to v1.7.1 (no public eMC API). 'global' queries US.one of
us · eu · uk · globalindicationstringOptional: narrow results to indications containing this text (e.g., 'pediatric', 'migraine').
force_refreshbooleanBypass 24h cache and re-fetch from primary source. Use for same-day label changes.default: false
include_label_historybooleanReserved for v1.7.2 — ignored in v1.7.0.default: false
governance.self_checkScore an AI-assisted HTA/HEOR workflow against 6 governance dimensions (transparency, citation validation, human-in-the-loop, PHI/data handling, bias & equity, auditability), each traced to the verified ELEVATE-GenAI reporting domains (ISPOR Working Group on Generative AI, Val...6 params
Score an AI-assisted HTA/HEOR workflow against 6 governance dimensions (transparency, citation validation, human-in-the-loop, PHI/data handling, bias & equity, auditability), each traced to the verified ELEVATE-GenAI reporting domains (ISPOR Working Group on Generative AI, Val...
Parameters* required
modestring'describe' = score a workflow_description + probes; 'audit_record' = auto-derive from a HEORAgent AuditRecord.one of
describe · audit_recorddefault: describeprobesobjectOne probe per dimension. Each is {answer: 'yes'|'no'|'partial'|null, detail?}. Omit or null → scored 'insufficient' (never a pass).
contextobjectaudit_recordobjectA prior HEORAgent AuditRecord (audit_record mode). Auto-derives transparency/citation/auditability.
ai_disclosure_levelstringAI assistance disclosure level appended to the output. "off" = no disclosure (analyst scratch mode); "standard" = default visible block (model, tools called, sources, date, human-review reminder); "submission" = standard + ISPOR ELEVATE-GenAI citation (use for regulatory or payer submissions).one of
off · standard · submissionworkflow_descriptionstringFree-text description of the AI-assisted workflow: what the AI does and where its output goes.
data.claims_queryQuery real-world claims and survey data across 14 datasets spanning the US and Latin America. US datasets: meps (expenditure survey, 2017-2023, costs in USD), namcs (physician office visits 2018-2022), nhamcs_ed (ED visits 2011-2022), ny_sparcs (NY inpatient 2017-2024, costs i...11 params
Query real-world claims and survey data across 14 datasets spanning the US and Latin America. US datasets: meps (expenditure survey, 2017-2023, costs in USD), namcs (physician office visits 2018-2022), nhamcs_ed (ED visits 2011-2022), ny_sparcs (NY inpatient 2017-2024, costs i...
Parameters* required
sexstringone of
male · female · alldefault: alltop_nintegerdefault: 20
age_maxintegerage_minintegerregionsarrayRegion/state filter. Brazil datasus UF codes: ['SP'] = São Paulo, ['SP','RJ','MG'] = top 3 states. US NAMCS/MEPS census regions: ['Northeast','South','West','Midwest']. Omit for all regions.
year_tointegerdatasetsarrayyear_fromintegermeps: 2017-2023 | ny_sparcs: 2017-2022, 2024 (no 2023) | nhamcs_ed: 2011-2022 | nhanes: 1999-2021 | nhis: 2019-2023 | ecuador_inec: 2022-2023 | uruguay_eh: 2016-2024 | mexico_egresos: 2018-2025 | colombia_rips: 2018-2023 | brazil_datasus/datasus_sih: 2018-2024 | chile_deis: 2018-2024 | namcs: 2018-2022
drug_namesarraySubstring match in drug list. E.g. ['metformin'] or ['empagliflozin', 'dapagliflozin'].
aggregationstringprevalence or count: record counts + weighted estimates by year — USE THIS for population sizing and budget impact. drug_utilization: top drugs by frequency. demographics: age/sex breakdown. comorbidities: top co-diagnosis ICD codes. cost: cost distribution (BRL for datasus, USD for meps/ny_sparcs).one of
prevalence · count · drug_utilization · demographics · comorbidities · costicd10_prefixesarrayICD-10 prefixes: ['E11'] = T2D, ['I21'] = AMI, ['C50'] = breast cancer, ['J45'] = asthma.
data.query_agentValidated epidemiology agent. Use this INSTEAD of data.claims_query when you need population-level estimates for budget impact models or population sizing. Automatically resolves clinical indications to ICD-10 codes, selects the right dataset by country, runs the query, valida...10 params
Validated epidemiology agent. Use this INSTEAD of data.claims_query when you need population-level estimates for budget impact models or population sizing. Automatically resolves clinical indications to ICD-10 codes, selects the right dataset by country, runs the query, valida...
Parameters* required
sexstringone of
male · female · alldefault: allage_maxintegerage_minintegerpurposestringDrives how the recommended_estimate is framed in the output.one of
population_sizing · budget_impact · epidemiologydefault: population_sizingregionsarrayOptional region filter. Brazil UF codes: ['SP','RJ','MG']. US census: ['Northeast','South'].
year_tointegeryear_fromintegerindicationstringClinical indication in plain English: 'type 2 diabetes inpatient', 'heart failure', 'COPD', 'breast cancer'. Maps to ICD-10 automatically.
country_codesarrayISO-2 country codes: ['BR'], ['US'], ['MX'], ['EC'], ['UY'], ['CO'], ['AR'].
additional_icd_prefixesarrayOverride or supplement auto-resolved ICD-10 prefixes. Use when the indication isn't in the built-in map.
HEORAgent MCP Server
AI-powered Health Economics and Outcomes Research (HEOR) agent as a Model Context Protocol server.
Try it now → HEORAgent on ChatGPT (ChatGPT Plus / Team) · Web UI (Claude, BYOK) ·
npx heor-agent-mcpfor Claude Desktop / Claude Code
Automates literature review across 44 data sources, risk of bias assessment (RoB 2 / ROBINS-I / AMSTAR-2), EQ-5D value set impact estimation, state-of-the-art cost-effectiveness modelling, HTA dossier preparation for NICE / EMA / FDA / IQWiG / HAS / EU JCA, and a persistent project knowledge base — all callable as MCP tools from Claude.ai, Claude Code, and any MCP-compatible host.
Built for pharmaceutical, biotech, CRO, and medical affairs teams who need rigorous, auditable HEOR workflows without building infrastructure from scratch.
First 60 seconds
Verify your install works before wiring it into Claude / Cursor / Continue. Open two terminal tabs:
Tab 1 — start the server in HTTP mode:
MCP_HTTP_PORT=8080 npx heor-agent-mcp@latest
You should see:
HEORAgent MCP server running on HTTP port 8080
Tab 2 — confirm it responds:
curl -s http://localhost:8080/health
Expected output:
{"status":"ok","server":"heor-agent-mcp","version":"1.10.2"}
✅ If you see the JSON above, the npm package works on your machine. Any further issues are in your MCP client config (Claude Desktop / Cursor / Continue), not the server.
❌ If you see command not found, run node --version — you need Node ≥20. If you see a different error, file a quick issue at https://github.com/neptun2000/heor-agent-mcp/issues with the output.
Now stop Tab 1 (Ctrl+C) and pick your client below — you don't need the HTTP mode for the actual integration; Claude / Cursor / Continue all use stdio.
Quick Start (per client)
Pick your MCP host:
Claude Code
claude mcp add heor-agent -- npx heor-agent-mcp
Then restart Claude Code.
Claude Desktop / claude.ai Desktop
Edit your MCP config file (~/Library/Application Support/Claude/claude_desktop_config.json on macOS) and add:
{
"mcpServers": {
"heor-agent": {
"command": "npx",
"args": ["heor-agent-mcp"]
}
}
}
Then restart Claude Desktop.
Cursor / Continue / Cline
Same config shape as Claude Desktop above; the file path differs by client:
- Cursor:
Settings → MCP → Add new MCP server - Continue:
~/.continue/config.jsonunder themcpServerskey - Cline:
Settings → MCP Servers → Edit MCP Settings
Hosted (no install)
- ChatGPT (Plus / Team): HEORAgent on ChatGPT — type
/heorto use it; works on any conversation. - Web UI (Claude, BYOK): web-michael-ns-projects.vercel.app — bring your Anthropic API key; runs the full v1.6.3 toolset.
Your first prompt
Once your MCP host is configured, paste any of these to verify end-to-end:
Run a literature search for semaglutide cost-effectiveness in T2D
using PubMed, NICE TAs, and ICER reports. Set runs=2.
Run irb_review for an industry-funded interventional Phase 2 trial in
relapsed MM — multi-site US+EU, pseudonymized data, greater-than-minimal
risk. I need the review tier, GDPR/HIPAA DMP, SAE framework, and the
ready-to-paste cover letter.
Run jca_pico_scope for osimertinib in EGFR-mutant 2L NSCLC across
DE/FR/IT/ES/NL. Then prepare an EU JCA dossier draft using the picos.
The first prompt exercises literature_search + validate_links (free, no API keys needed). The second exercises irb_review (pure decision tree, instant). The third exercises jca_pico_scope → hta_dossier pipeline.
What's new
See CHANGELOG.md for full version history. Current: v1.23.0 (45 tools, 44 data sources).
v1.17.0–v1.23.0 — Living Evidence Intelligence (review → reimbursement)
A connected RWE + cross-deliverable layer (see docs/FEATURES.md for the full table):
- RWE & real-world safety:
rwe.method_select(study-design selection),pv.comparative_safety(class-level FAERS-style AE ranking),evidence.triangulation(per-outcome RCT↔RWE concordance). - Governed social listening:
rwe.social_listening_protocol+pv.social_listening_triage(GVP Module VI ICSR triage — no scraping). - One source of truth:
evidence.claim_registry(author/auto-import a figure once),evidence.consistency_check(detect drift across dossier/publication/payer),publication.draft(reuse claims; CONSORT/STROBE/PRISMA/CHEERS + GPP2022/ICMJE). - Living orchestration:
evidence.gap_analysis(iEGP),workflow.living_evidence(SLR → living KB → JCA/HTA runbook),hta.living_gvd(regenerate only the GVD sections whose figures changed).
v1.13.0 — AI Transparency Disclosure (ISPOR ELEVATE-GenAI aligned)
16 tools now accept an ai_disclosure_level parameter:
| Value | Behaviour |
|---|---|
"off" | No disclosure block appended |
"standard" | Model ID · tools called · data sources · date · human-review reminder |
"submission" | Standard block + ISPOR ELEVATE-GenAI full citation |
Default by tool tier: HTA/regulatory tools (hta_dossier, hta_workflow, jca_pico_scope, pv_classify, etc.) default to "submission"; analysis tools (risk_of_bias, cost_effectiveness_model, etc.) default to "standard". Pass ai_disclosure_level: "off" to suppress.
Environment-level default: set HEORAGENT_DISCLOSURE_LEVEL=off|standard|submission to override the built-in per-tool defaults globally.
Web UI persona defaults: payer and HTA-reviewer personas always use "submission"; analyst personas default to "standard" and switch to "off" for scratch / exploratory prompts.
v1.0.4 highlights (still in v1.6.3)
Pharmacovigilance + workflow orchestration:
pv_classifytool — classifies a planned study into its EMA pharmacovigilance regulatory category (PASS imposed/voluntary, PAES, RMP Annex 4, DUS, active surveillance registry, pregnancy registry, spontaneous reporting, ICH E2E plan). Returns the matching GVP module (V/VI/VIII/VIII Addendum I), ENCePP protocol template ID, RMP implications, FDA analogue, and submission obligations. Pure decision-tree per EMA GVP rev 4 + EU Regulation 1235/2010 Article 107a. <200ms response.hta_dossierPharmacovigilance Plan section — passpv_classificationfrompv_classifytohta_dossierand the dossier output now includes a PV Plan section between RoB and CEA. Without it, a one-line "PV plan not provided" note flags the gap so reviewers see what's missing.maic_workfloworchestrator (v1.0.6) — runs the full MAIC discovery+screening pipeline (ITC feasibility + parallel literature_search + screening + RoB + network) in one MCP call. Built for ChatGPT-5.3 surfaces where chaining 5+ tool calls in parallel is unreliable; works equally well from Claude.examplestool (v1.0.5) — pre-filled JSON inputs for heavy-schema tools (CEA, BIA, survival, MAIC, Bucher) plus amaic_workflow_recipemulti-step prompt template for ChatGPT users.- CMS IRA awareness — when
pv_classifyis called with US jurisdiction, output explicitly notes that CMS IRA Medicare price-negotiation calculations exclude PV cost data — track those obligations in the regulatory budget, not the HEOR cost-effectiveness model. - GRADE I²-based inconsistency, GRADE upgrading (Guyatt 2011), Bucher consistency check, EQ-5D 5L baseline-utility-aware impact (v1.0.4) — see CHANGELOG.md.
- ChatGPT Custom GPT support (v1.0.4) — OpenAPI 3.1 adapter at
/api/openapilets you build a Custom GPT in 5 minutes. See ChatGPT Custom GPT below. - Surface-tagged analytics (v1.0.4) — every
tool_callPostHog event carries asurfaceproperty (claude_anthropic_web/chatgpt_adapter/claude_desktop/direct_mcp).
See CHANGELOG.md for the full diff.
Tools (28)
| Tool | Purpose |
|---|---|
literature_search | Search 44 data sources with a full PRISMA-style audit trail |
screen_abstracts | PICO-based relevance scoring and study design classification |
risk_of_bias | Cochrane RoB 2 / ROBINS-I / AMSTAR-2 with GRADE RoB domain summary |
evidence_network | Build treatment comparison network and assess NMA feasibility |
evidence_indirect | Bucher and frequentist NMA with automatic consistency check vs direct h2h evidence (NICE DSU TSD 18) |
population_adjusted_comparison | MAIC and STC for population-adjusted indirect comparisons |
survival_fitting | Fit 5 parametric distributions to KM data (NICE DSU TSD 14) |
itc_feasibility | Assess the 3-assumption ITC framework and recommend Bucher / NMA / MAIC / STC / ML-NMR |
cost_effectiveness_model | Markov / PartSA / decision-tree CEA with PSA, OWSA, CEAC, EVPI, EVPPI; QALY + evLYG support |
budget_impact_model | ISPOR-compliant BIA with year-by-year output and treatment-displacement modelling |
hta_dossier | Draft submissions for NICE, EMA, FDA, IQWiG, HAS, and EU JCA — GRADE table uses structured RoB when rob_results passed; inconsistency uses I² when heterogeneity_per_outcome passed; GRADE upgrading (Guyatt 2011) supported via upgrading_per_outcome |
utility_value_set | EQ-5D-3L / 5L value-set reference + baseline-utility-aware Biz 2026 ICER impact estimator (UK 5L transition) |
validate_links | HTTP validation of citation URLs before presentation |
project_create | Initialize a persistent project workspace |
knowledge_search | Full-text search across a project's raw/ and wiki/ trees |
knowledge_read | Read any file from a project's knowledge base |
knowledge_write | Write compiled evidence to the project wiki (Obsidian-compatible) |
literature_search
Searches across 44 sources in parallel. Every call returns a source selection table showing which of the 44 sources were used and why — essential for HTA audit trails.
Example call:
{
"query": "semaglutide cardiovascular outcomes type 2 diabetes",
"sources": ["pubmed", "clinicaltrials", "nice_ta", "cadth_reviews", "icer_reports"],
"max_results": 20,
"output_format": "text"
}
cost_effectiveness_model
Multi-state Markov model (default) or Partitioned Survival Analysis (oncology), following ISPOR good practice and NICE reference case (3.5% discount rate, half-cycle correction). Includes:
- PSA — 1,000–10,000 Monte Carlo iterations, probability cost-effective at WTP thresholds
- OWSA — one-way sensitivity analysis with tornado summary
- CEAC — cost-effectiveness acceptability curve
- EVPI — expected value of perfect information
- WTP assessment — verdict against NHS (£25–35K/QALY, updated April 2026), US payer ($100–150K), societal thresholds
Example call:
{
"intervention": "Semaglutide 1mg SC weekly",
"comparator": "Sitagliptin 100mg daily",
"indication": "Type 2 Diabetes Mellitus",
"time_horizon": "lifetime",
"perspective": "nhs",
"model_type": "markov",
"clinical_inputs": { "efficacy_delta": 0.5, "mortality_reduction": 0.15 },
"cost_inputs": { "drug_cost_annual": 3200, "comparator_cost_annual": 480 },
"utility_inputs": { "qaly_on_treatment": 0.82, "qaly_comparator": 0.76 },
"run_psa": true,
"output_format": "docx"
}
hta_dossier_prep
Drafts submission-ready sections for six HTA frameworks with gap analysis:
| Body | Country | Submission types |
|---|---|---|
| NICE | UK | STA, MTA, early_access |
| EMA | EU | STA, MTA |
| FDA | US | STA, MTA |
| IQWiG | Germany | STA, MTA |
| HAS | France | STA, MTA |
| JCA | EU (Reg. 2021/2282) | initial, renewal, variation (with PICOs) |
Accepts piped output from literature_search and cost_effectiveness_model.
risk_of_bias
Assesses risk of bias using the appropriate Cochrane instrument, auto-detected from study_type:
| Study type | Instrument |
|---|---|
| RCT | RoB 2 (5 domains: randomization, deviations, missing data, measurement, reporting) |
| Observational | ROBINS-I (7 domains: confounding, selection, classification, deviations, missing data, measurement, reporting) |
| Systematic review | AMSTAR-2 (16 items, critical vs non-critical) |
Returns a rob_results object you can pass directly to hta_dossier_prep — this replaces the heuristic RoB estimate in the GRADE table with structured domain judgments.
Example call:
{
"studies": [{ "id": "pmid_1", "study_type": "RCT", "title": "...", "abstract": "..." }],
"output_format": "json"
}
Pipeline:
literature_search→screen_abstracts→risk_of_bias→hta_dossier_prep
Knowledge base tools
Projects live at ~/.heor-agent/projects/{project-id}/ with:
raw/literature/— auto-populated literature search resultsraw/models/— auto-populated model runsraw/dossiers/— auto-populated dossier draftsreports/— generated DOCX fileswiki/— manually curated, Obsidian-compatible markdown with[[wikilinks]]
Pass project: "project-id" to any tool and results are saved automatically.
Examples
Copy-paste prompts to try in Claude Code, Claude Desktop, or the web UI.
Single-tool examples
Literature search
Search the literature for tirzepatide cardiovascular outcomes in type 2 diabetes. Use PubMed, ClinicalTrials.gov, and NICE TAs.
Survival curve fitting
Fit survival curves to this OS data from KEYNOTE-189: time 0 survival 1.0, time 6 survival 0.88, time 12 survival 0.72, time 18 survival 0.60, time 24 survival 0.51, time 36 survival 0.38. Use months.
Budget impact
Estimate the 5-year NHS budget impact of semaglutide for obesity. 200,000 eligible patients, drug cost £1,200/year, comparator (orlistat) £250/year, uptake 15% year 1 to 40% year 5.
Cost-effectiveness model
Build a CE model for semaglutide vs sitagliptin in T2D, NHS perspective, lifetime horizon, with PSA.
Indirect comparison (Bucher)
I have two trials: SUSTAIN-1 showed semaglutide vs placebo HR 0.74 (0.58-0.95) for HbA1c, and AWARD-5 showed dulaglutide vs placebo HR 0.78 (0.65-0.93). Run a Bucher indirect comparison between semaglutide and dulaglutide.
MAIC (population-adjusted comparison)
Run a MAIC between SUSTAIN-7 (N=300, semaglutide vs placebo, HR 0.74, CI 0.58-0.95, age 56±10, BMI 33±5) and AWARD-11 (N=600, dulaglutide vs placebo, HR 0.78, CI 0.65-0.93, age 58±9, BMI 35±6). Adjust for age and BMI.
Multi-tool workflows
Abstract screening workflow
Search PubMed for pembrolizumab in NSCLC, then screen the results with population adults with NSCLC, intervention pembrolizumab, comparator chemotherapy, outcomes overall survival and PFS.
Evidence network + NMA feasibility
Search for GLP-1 receptor agonists in T2D using PubMed, build an evidence network from the results, and assess NMA feasibility.
CE model with scenarios
Build a CE model for dapagliflozin vs placebo in heart failure, NHS perspective, lifetime horizon, with PSA. Add scenarios: "20% price reduction" with drug cost 400, "10-year horizon" with time_horizon 10yr.
End-to-end HTA workflow
Full dossier preparation
Create a project for semaglutide in obesity targeting NICE and ICER. Search literature for evidence, screen the results for adults with obesity comparing semaglutide to placebo for weight loss outcomes, assess risk of bias on the screened studies, then draft a NICE STA dossier using the screened results and rob_results.
This single prompt exercises: project_create → literature_search → screen_abstracts → risk_of_bias → hta_dossier_prep (GRADE RoB from structured assessment).
Data Sources
44 sources across 10 categories. Every literature_search call includes a source selection table showing used/not-used status and reason for each.
Biomedical & Clinical Trials (5)
- PubMed — 35M+ biomedical citations (NCBI E-utilities)
- ClinicalTrials.gov — NIH/NLM trial registry (CT.gov v2 API)
- bioRxiv / medRxiv — Life sciences and medical preprints
- ChEMBL — Drug bioactivity, mechanisms, ADMET (EMBL-EBI)
- Wiley Online Library — Pharmacoeconomics, Health Economics, Journal of Medical Economics, Value in Health (CrossRef, ~77% abstract coverage, no key required)
Epidemiology & Demographics (5)
- WHO GHO — WHO Global Health Observatory
- World Bank — Demographics, macroeconomics, health expenditure
- OECD Health — OECD health statistics (expenditure, workforce, outcomes)
- IHME GBD — Global Burden of Disease (DALYs, prevalence across 204 countries)
- All of Us — NIH precision medicine cohort
FDA (2)
- FDA Orange Book — Drug approvals and therapeutic equivalence
- FDA Purple Book — Licensed biologics and biosimilars
HTA Appraisals (10) — HTA precedent decisions
- NICE TAs (UK) · CADTH (Canada) · ICER (US) · PBAC (Australia)
- G-BA AMNOG (Germany) · IQWiG (Germany) · HAS (France)
- AIFA (Italy) · TLV (Sweden) · INESSS (Quebec, Canada)
HTA Cost References (5)
- CMS NADAC (US drug acquisition costs)
- PSSRU (UK unit costs) · NHS National Cost Collection · BNF (UK drug pricing)
- PBS Schedule (Australia)
LATAM (6)
- DATASUS · CONITEC · ANVISA (Brazil)
- PAHO (Pan American regional) · IETS (Colombia) · FONASA (Chile)
APAC (1)
- HITAP (Thailand)
Enterprise (6) — require API key
| Source | Env variable |
|---|---|
| Embase | ELSEVIER_API_KEY |
| ScienceDirect | ELSEVIER_API_KEY |
| Cochrane Library | COCHRANE_API_KEY |
| Citeline | CITELINE_API_KEY |
| Pharmapendium | PHARMAPENDIUM_API_KEY |
| Cortellis | CORTELLIS_API_KEY |
| Google Scholar | SERPAPI_KEY |
HEOR Methodology & Utility Reference (3)
- ISPOR — HEOR methodology and conference abstracts
- OHE (Office of Health Economics) — EQ-5D value set research and HEOR methodology
- EuroQol Group — EQ-5D instruments, value sets, and registry
Output Formats
All tools support output_format:
text(default) — Markdown with formatted tables and headingsjson— Structured objects for downstream toolsdocx— Microsoft Word document, saved to disk, path returned in response
DOCX files are saved to ~/.heor-agent/projects/{project}/reports/ (when a project is set) or ~/.heor-agent/reports/ (global). The tool response contains the absolute path — ready to attach to submissions or share with stakeholders.
Audit Trail
Every tool call returns a full audit record:
- Source selection table — all 44 sources with used/not-used and reason
- Sources queried — queries sent, response counts, status, latency
- Inclusions / exclusions — counts with reasons
- Methodology — PRISMA-style for literature, ISPOR/NICE for economics
- Assumptions — every assumption logged with justification
- Warnings — data quality flags, missing API keys, failed sources
Suitable for inclusion in HTA submission appendices.
Configuration
# Optional — enterprise data sources
ELSEVIER_API_KEY=... # Embase + ScienceDirect
COCHRANE_API_KEY=... # Cochrane Library
CITELINE_API_KEY=... # Citeline
PHARMAPENDIUM_API_KEY=... # Pharmapendium
CORTELLIS_API_KEY=... # Cortellis
SERPAPI_KEY=... # Google Scholar
# Optional — knowledge base location
HEOR_KB_ROOT=~/.heor-agent # Default
# Optional — localhost proxy for enterprise APIs behind corporate VPN
HEOR_PROXY_URL=http://localhost:8787
# Optional — hosted tier (future)
HEOR_API_KEY=...
Web UI
A companion chat interface is available at:
https://web-michael-ns-projects.vercel.app
- Chat with Claude Sonnet 4.6 + all 22 HEOR tools
- BYOK (Bring Your Own Key) — paste your Anthropic API key in the settings; it stays in your browser's localStorage and is never stored on our servers
- Markdown rendering with styled tables, tool call cards with live progress timers, and theme-aware mermaid network diagrams
- 12 example prompts covering literature search, CEA, BIA, NMA, ITC feasibility, RoB, EQ-5D 5L, EU JCA dossiers
- Per-request MCP sessions (no cross-user session bleed)
The web UI calls the hosted MCP server on Railway for tool execution. No setup required — just add your API key and start querying.
Self-hosting the web UI
cd web
npm install
echo "ANTHROPIC_API_KEY=sk-ant-..." > .env.local # optional server-side fallback
npm run dev -- -p 3456
Set MCP_SERVER_URL to point to your own MCP server instance (default: the public Railway deployment).
ChatGPT Custom GPT
🟢 Live: HEORAgent on ChatGPT →
Open in ChatGPT (Plus / Team / Enterprise account required), pick a conversation starter, and you're querying 44 HEOR data sources.
HEORAgent is also available as a ChatGPT Custom GPT — useful when you (or your team) prefer the ChatGPT interface or have a ChatGPT Plus/Team account but no Anthropic API access.
Behind the scenes, the web tier exposes an OpenAPI 3.1 adapter at /api/openapi, with one POST endpoint per tool at /api/v1/{tool_name}. ChatGPT speaks this contract natively.
What's different from the Anthropic surface
| Web UI / MCP / Claude Desktop | ChatGPT Custom GPT | |
|---|---|---|
| Streaming | yes (SSE) | no (45s single response) |
psa_iterations | up to 10,000 | capped to 1,000 (CEA) / 500 (BIA) |
literature_search.runs | 1–5 | capped to 1 |
literature_search.max_results | up to 100 | capped to 30 |
| Auth model | BYOK Anthropic | optional X-API-Key header (server-side CHATGPT_ADAPTER_TOKEN) |
| Surface label in PostHog | claude_anthropic_web / claude_desktop | chatgpt_adapter |
The caps exist because ChatGPT Actions hard-fail at the 45-second response timeout. PSA, multi-run literature search, and full max_results would routinely exceed it. The web UI and MCP clients are unaffected.
Build a Custom GPT (ChatGPT Plus / Team required)
- Visit chatgpt.com/gpts/editor and click Create.
- Configure tab — fill in name (e.g., "HEORAgent"), description, and conversation starters. Paste the system prompt from
web/lib/claude.ts(or write your own — the tool descriptions are self-documenting). - Actions → Create new action → Import from URL → paste:
ChatGPT auto-imports all 17 endpoints with their schemas.https://web-michael-ns-projects.vercel.app/api/openapi - Authentication — choose None for the open public endpoint, or API Key with the
CHATGPT_ADAPTER_TOKENvalue if you've configured one (recommended for prod). - Privacy policy URL — required by GPT Store. Use the web UI's privacy URL or your own.
- Test in the playground (right pane), then Publish → "Anyone with the link" or "GPT Store".
Securing the adapter for production
By default the /api/v1/* endpoint is open. Two layers of protection are recommended for any public-facing GPT:
# 1. Token-gate the endpoint
cd web
vercel env add CHATGPT_ADAPTER_TOKEN production # generate a long random token
# Configure the same token in your Custom GPT under Authentication → API Key
# 2. Built-in rate limit
# 60 req/min per IP is enforced automatically (lib/rateLimit.ts).
# For multi-region/high-traffic prod, swap in @upstash/ratelimit + Vercel KV.
Sample call (manual, no GPT needed)
curl -X POST https://web-michael-ns-projects.vercel.app/api/v1/utility_value_set \
-H "Content-Type: application/json" \
-d '{
"action": "estimate_impact",
"indication_type": "non_cancer_qol_only",
"baseline_utility": 0.85,
"base_icer": 30000
}'
Returns the Biz 2026 baseline-utility-adjusted ICER projection (the new EQ-5D 5L impact estimator).
HTTP Transport
The server supports both stdio (default, for local MCP clients) and Streamable HTTP (for hosted deployment).
# Stdio mode (default — for Claude Code, Claude Desktop)
npx heor-agent-mcp
# HTTP mode — for hosted deployment, Smithery, web UI backend
npx heor-agent-mcp --http # port 8787
MCP_HTTP_PORT=3000 npx heor-agent-mcp # custom port
HTTP endpoints:
POST/GET/DELETE /mcp— MCP Streamable HTTP protocolGET /health— health checkGET /.well-known/mcp/server-card.json— Smithery discovery
Development
git clone https://github.com/neptun2000/heor-agent-mcp
cd heor-agent-mcp
npm install
npm test # 401 tests across 84 suites
npm run build # Compile TypeScript to dist/
npm run dev # Run with tsx (no build step)
Requires: Node.js ≥ 20.
Architecture
┌────────────────────────────────────────────┐
│ MCP Host (Claude.ai / Claude Code / etc.) │
└────────────────┬───────────────────────────┘
│ stdio
┌────────────────▼──────────────────────────┐
│ heor-agent-mcp server │
│ ┌──────────────────────────────────────┐ │
│ │ 17 MCP tools (Zod-validated) │ │
│ ├──────────────────────────────────────┤ │
│ │ DirectProvider (default) │ │
│ │ ├─ 44 source fetchers │ │
│ │ ├─ Audit builder + PRISMA trail │ │
│ │ ├─ Markov / PartSA economic models │ │
│ │ ├─ Markdown + DOCX formatters │ │
│ │ └─ Knowledge base (YAML + MD) │ │
│ └──────────────────────────────────────┘ │
└───────────────────────────────────────────┘
│
┌────────────┴─────────────┐
▼ ▼
┌────────────┐ ┌──────────────────┐
│ ~/.heor- │ │ External APIs │
│ agent/ │ │ (PubMed, NICE, │
│ projects/ │ │ ICER, CADTH, …) │
└────────────┘ └──────────────────┘
License
MIT — see LICENSE.
Trust & Transparency
HEORAgent is a research and analysis tool — not a clinical decision-support system. It is not classified as high-risk under the EU AI Act because it does not drive individual diagnosis, treatment, or monitoring; it falls under limited-risk transparency obligations only. Every output is intended for review by a qualified HEOR/HTA/PV professional before any action is taken.
- EU AI Pact signatory — committed to AI governance, high-risk system mapping, and AI literacy promotion (voluntary commitments per the European Commission, ahead of the AI Act's August 2026 deadline).
- PRISMA-style audit trail on every tool call (sources queried, succeeded, failed, assumptions applied).
- AI commentary explicitly labelled — domain claims (ICERs, trial results, regulatory decisions) come exclusively from tool outputs, never from training-data recall.
- Methodology cited inline — ISPOR, NICE DSU TSDs, NICE PMG36, Cochrane Handbook, GRADE, EMA GVP, Cope 2014, Phillippo 2016, Biz 2026.
Full statement: /ai-transparency — risk classification, human oversight model, methodological references, and reporting channel.
Disclaimer
All outputs are preliminary and for research orientation only. Results require validation by a qualified health economist before use in any HTA submission, payer negotiation, regulatory filing, or clinical decision. This tool does not replace professional HEOR expertise.
Distribution
| Channel | How to use | Who pays |
|---|---|---|
| npm | npx heor-agent-mcp | User's Claude subscription |
| Smithery | smithery.ai/servers/neptun2000-70zu/heor-agent-mcp | User's Claude subscription |
| Web UI | web-michael-ns-projects.vercel.app | User's own Anthropic API key (BYOK) |
| Hosted MCP | https://heor-agent-mcp-production.up.railway.app | Free (tool execution only) |
Links
Featured
Keep your Mac awake while Claude Code and 40+ AI agents run. Sleeps when they're idle.
One time payment $9 →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →On Capafy, your Skill runs online 24/7 as an agent product, and you get paid every time someone uses it.
Start earning →Categories
Registryactive
Packageheor-agent-mcp
TransportSTDIO
UpdatedJun 6, 2026
