Skip to main content
Glama
Sign In
SINAN MCP

SINAN MCP

by miscrapha-DELTA-IA
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

SINAN MCP

Open-source reference discovery and reproducible research workflows for Brazilian notifiable diseases

Documentação em português · Guia rápido · Scientific evidence · External review · Security

CI · MIT License · Python 3.11+

IMPORTANT

Independent research prototype. Not operated, certified or endorsed by Brazil's Ministry of Health, SINAN or DATASUS. The public MCP service returns fictional demonstration records only, not official case counts, patient data or validated clinical indicators.

Overview

SINAN MCP uses the Model Context Protocol (MCP) to provide structured access to curated public-health documentary references. It keeps documentary research, fictional software demonstrations and restricted local analysis in separate environments.

Component

Available today

Boundary

Reference catalogue

14 internal condition aliases and 17 curated institutional reference pages

A reference link does not establish access to a usable dataset

Zika/TB documentary evidence

Seven official PDFs with locked SHA-256; 57 matrix records from 3 dictionaries, reused as 42 Zika + 20 TB bundle entries

All 57/57 code occurrences were rechecked under stricter identifier boundaries in CI 42144137431; neither semantics nor the 2024 export layout is validated

Public MCP preview

10 read-only tools, including 120 deterministic fictional TB demonstration records

No actual SINAN notifications, real indicators, SQLite or external data import

Local research tools

Experimental ingestion, diagnostics and methodological utilities

Not part of the public service; real-data work requires separate authorization and controls

Related MCP server: CDC MCP

Architecture

flowchart LR
  A["MCP client"] --> B["Public read-only service"]
  B --> C["Source catalogue and documentary metadata"]
  B --> D["Fictional TB demonstration"]

  E["Controlled research environment"] --> F["Local tools and datasets"]
  F --> G["Scientific and privacy review"]
  G --> H["Potential separately approved outputs"]

The public service has no path to the private research database. Passing software tests does not approve epidemiological estimates or the release of personal data.

Documentary validation

The Zika and tuberculosis evidence dossier details document identity, apparent versions, byte-integrity checks and page-level field references.

  • Established historically: 7/7 PDF byte-integrity checks and 57/57 text occurrences under the previous verifier. The tightened identifier test passed CI 42144137431; occurrence is never semantic validation.

  • Pending: independent human review of interpretations, 2024 dataset-layout concordance and privacy aspects.

  • Not authorized: publication of real SINAN counts, individual-level access or medical decisions.

External contributors can use the review guide. Scientific tasks and decisions are tracked in issue #62, separate from the presentation documents.

Public synthetic preview

Resource

Address

MCP endpoint (for compatible clients)

/mcp

Health status

/health

About and support

About · Support

Policies

Privacy · Terms

The /mcp endpoint is a machine-to-machine protocol, not a normal web page. Free hosting may introduce startup delays and does not guarantee continuous availability.

Local demonstration

Requires Python 3.11+ and Git. The commands below generate synthetic fixtures only.

git clone https://github.com/miscrapha-DELTA-IA/SINAN-MCP.git
cd SINAN-MCP
py -3.11 -m venv .venv
.\.venv\Scripts\python.exe -m pip install -e ".[dev]"
.\.venv\Scripts\python.exe -m sinan_mcp demo
.\.venv\Scripts\python.exe -m pytest -q

See installation instructions for other platforms and the technical operations index.

Documentation and contribution

Read

Purpose

Documentation centre

Concise navigation in Portuguese

Public quickstart

Glossary, remote MCP connection and fictional example prompts

Research reference library

Disease coverage, source provenance and methods

Zika/TB evidence dossier

Primary documents and scientific limitations

Privacy and governance

Data handling and unresolved legal questions

Operating and integration guides

Deployment, clients, testing and security

Contributing

Reproducible reports and synthetic-only contributions

Changelog

Versioned software history

Original source code is distributed under the MIT License. This does not convey ownership of official documents or government trademarks.

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityMaintained
ResponsivenessNo recent community issues

Resources

Related MCP Connectors

Related MCP Servers

  • CDC MCP

    pipeworx-io
    A
    license
    Not graded
    quality
    B
    maintenance
    Enables AI agents to query U.S. public-health datasets such as disease surveillance, mortality, behavioral risk factors, and flu activity through the Socrata API, using natural language or direct tool calls.
    311 npm
    MIT

  • who-tb

    pipeworx-io
    A
    license
    Not graded
    quality
    C
    maintenance
    Enables querying WHO Global Tuberculosis Programme data, including TB burden estimates, case notifications, and programme financing across 217 countries from 2000 onward.
    220 npm
    MIT

  • saudeemdado-mcp

    pedropaulofernandes88-stack
    A
    license
    Not graded
    quality
    C
    maintenance
    Enables natural-language querying of Brazilian public health indicators from DataSUS microdata — mortality, dengue, hospitalizations, ICSAP, and births — through 41 tools covering consultation, analysis (peer-stratum comparison, dengue endemic channel, weekly bulletin), reliability warnings, and anomaly detection. It runs locally against the public API with anti-hallucination rules so every figure comes from a tool with its source cited.
    MIT