LABSCO SUMMARY
Drug and protocol answers with a published source behind them.
The value is the grounding rather than the retrieval: a branded-drug lookup returns composition and manufacturer from a curated corpus, and a protocol answer traces back to the body that published it — which is what a general model cannot supply for the Indian market.
What it is
A connector to medical knowledge bases curated for the Indian healthcare context, built to fix a specific gap: India has no centralised public repository of branded medications, so general models answer drug questions without anything underneath them. The tools are curated by eka.care's in-house doctors and validated against an internal questionnaire.
What you get
- Indian branded drug search — look up across 500,000+ branded drugs by name or generic composition, narrowed by drug form and volume, with generic composition and manufacturer returned
- Indian pharmacology details — a generic composition's details against the 2011 guidelines published by the National Formulary of India
- Indian treatment protocol search — standardised guidelines and decision pathways covering 175 medical conditions across 180 treatment protocols from bodies such as ICMR and RSSDI, retrieved by condition, then by publisher, then by query
Requirements
A developer key from eka.care — a client ID and client secret from console.eka.care, passed as --client-id and --client-secret alongside --eka-api-host in the client config. The code is open source but the APIs behind it are not: valid credentials are required, free access was offered initially, and the project says API usage will be charged in future. Install uv and run through uvx eka_mcp_server; if the client reports spawn uvx ENOENT, use the full path from which uvx as the command. uv cache clean eka_mcp_server forces a newer version when an old one is pinned in the cache. The project states plainly that this is information rather than medical advice, that clinical judgment remains the professional's, and that completeness is not guaranteed for newly approved medications or recently updated guidelines.
Setup effort
One command plus credentials — uvx eka_mcp_server, then supply credentials
USE CASES
01
Identifying what a branded medication actually contains
Search by the brand name and get the generic composition and manufacturer back, narrowed by form and volume.
02
Finding the brands behind a generic
Search by composition instead, to see which branded drugs available in India carry it.
03
Checking the guideline for a condition
Confirm the condition is one of the supported tags, find which bodies have published on it, then pull the relevant protocol content.
04
Grounding an answer in a named publisher
Protocols come from bodies such as ICMR and RSSDI, so the answer can be attributed rather than asserted.
05
Reading pharmacology from the national formulary
Look up a generic composition's details against the 2011 National Formulary of India guidelines.
BEFORE YOU INSTALL
What to know first
Requires
uvx on PATH
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
Server setup
TERMINAL
uvx eka_mcp_serverMCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"eka-mcp-server": {
"command": "uvx",
"args": [
"eka_mcp_server",
"--eka-api-host",
"https://api.eka.care",
"--client-id",
"<YOUR_CLIENT_ID>",
"--client-secret",
"<YOUR_CLIENT_SECRET>"
]
}
}
}