Open Masala MCP server
An ancestry-adjusted health reference oracle over the Open Masala dataset, exposed via the Model Context Protocol so any MCP client — Claude Desktop, Claude Code, an agent — can query cited, provenance-tiered reference values as tool calls.
Design principle: it holds no user data
The server is a stateless reference oracle. Tools like evaluate_value take a value as an argument and return a computed, cited result — nothing is stored, and no personal data ever lives here. All handling of a person's actual labs happens in the client / advisor layer (see ../advisor/), never in this server. That's what makes it safe to publish and run anywhere.
Tools
| Tool | What it does |
|---|---|
list_measures() |
Every analyte/measure in the dataset — the coverage map. Start here. |
get_reference(analyte) |
Ancestry-adjusted range(s)/threshold(s) for an analyte, paired with the general-population value, provenance tier, grade, citations, and overclaim guard. Matches a LOINC code, id, or name. |
evaluate_value(analyte, value, sex?, age?) |
Compare a person's own value against the South-Asian-adjusted threshold; returns a flag status + cited rationale + overclaim guard. Stateless. |
screening_for(age, sex?) |
Which SA-specific screenings/thresholds apply to a person of this age/sex. Cited. |
search(query) |
Free-text search across analytes, notes, guards, and sources. |
list_sources() |
Every unique citation backing the dataset. |
Resources
open-masala://dataset— the full canonical JSON.open-masala://guidance— the provenance-tier + overclaim usage rules a client should follow.
Install & run
No clone (recommended) — uv fetches and runs it from PyPI:
uvx open-masala-mcp
From a source checkout:
pip install "mcp[cli]"
python3 mcp-server/server.py # or: python -m open_masala_mcp.server
Add to Claude Desktop
Edit claude_desktop_config.json (macOS: ~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"open-masala": {
"command": "uvx",
"args": ["open-masala-mcp"]
}
}
}
Restart Claude Desktop; the Open Masala tools appear in the tools menu. Now you can ask, e.g., "I'm a South Asian man, 34, BMI 24 — what does Open Masala say?" and Claude will call evaluate_value / screening_for and answer with cited, guard-respecting reference values.
(From a checkout instead of PyPI, use "command": "python3", "args": ["/absolute/path/to/open-masala/mcp-server/server.py"].)
Add to Claude Code
claude mcp add open-masala -- uvx open-masala-mcp
Responsible-use contract
Every tool response includes usage_rules and a disclaimer. Clients (and the LLMs driving them) should:
- Report the
provenance_tier— never present astudy-derivedvalue as guideline-backed, and tell the user not to apply acontested-deprecatedadjustment. - Honor the
overclaim_guardon every row. - Cite the sources attached to each row.
- Defer diagnosis and treatment to a clinician.
This is a reference dataset, not a medical device.
Metadata
Release files for open-masala-mcp 0.0.3
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| open_masala_mcp-0.0.3.tar.gz | 24.4 kB | Without provenance | Details for open_masala_mcp-0.0.3.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| open_masala_mcp-0.0.3-py3-none-any.whl | Python 3 | none | any | Details for open_masala_mcp-0.0.3-py3-none-any.whl |
Total release size: 48.5 kB
Release files / open_masala_mcp-0.0.3.tar.gz
| Download URL | open_masala_mcp-0.0.3.tar.gz |
|---|---|
| Size | 24.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
af0f3dd09af134f08a89b66fb4292fc8ad13f2223f397a310e0baf0f1a1e7770
|
|
BLAKE2b-256 checksum How to use checksums |
f7276aa09e62c2e701b1919b989f4a87e973c354f8c0184fa549f9937726899d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.11.28 {"installer":{"name":"uv","version":"0.11.28","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Release files / open_masala_mcp-0.0.3-py3-none-any.whl
| Download URL | open_masala_mcp-0.0.3-py3-none-any.whl |
|---|---|
| Size | 24.1 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
6adeca05aa3fe646bbd6f3a02765c74dc38270ce8f1bb4d52cfe0ed2e6b8765d
|
|
BLAKE2b-256 checksum How to use checksums |
24a90c4403fb0629e67c40fc08ecfcb26c9ddc10a50718ceac6cf024ee427432
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.11.28 {"installer":{"name":"uv","version":"0.11.28","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Security for open-masala-mcp 0.0.3
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.