PREGNA-RISK: Pregnancy Risk in SLE/APS
v1.0.0Calculates a weighted composite score predicting adverse pregnancy outcomes in SLE/APS patients to guide risk stratification and management.
⭐ 0· 93·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name and description match the included code and instructions: a weighted composite risk score with Monte Carlo uncertainty estimation for SLE/APS pregnancy risk. The code implements the weights, scoring, classification, and reporting described in SKILL.md. One caution: registry metadata lists source/homepage as unknown — the package contains an implementation, but provenance and authorship claims (affiliations) are not verifiable from the registry entry.
Instruction Scope
SKILL.md only instructs installing numpy and running the local script; the script computes scores and prints a report. There are no instructions to read arbitrary system files, access environment variables, or send data to external endpoints.
Install Mechanism
Install spec is a single pip dependency (numpy). This is proportionate for Monte Carlo simulations. There is no download-from-URL, no extraction of remote archives, and no creation of nonstandard binaries.
Credentials
The skill requires no environment variables, no credentials, and no config paths. That is appropriate for a local calculator that performs purely local computation.
Persistence & Privilege
The skill is not always-enabled and requests no elevated or persistent privileges. It does not modify other skills or system settings.
Assessment
This code appears to be a straightforward, local risk calculator (no network calls or secret access). Before using it clinically: 1) review the full source (you have pregna_risk.py) to confirm behavior and to ensure no hidden I/O; 2) validate the model and weights against the cited literature and against real/known test cases; 3) run it in an isolated environment (virtualenv) and install numpy from PyPI; 4) do not feed identifiable patient data until you are satisfied with privacy practices — the tool prints reports to stdout and does not appear to exfiltrate data, but provenance is unclear (registry shows no homepage), so confirm authorship/maintenance and licensing; 5) treat outputs as decision support only and consult specialists before relying on the score for clinical care.Like a lobster shell, security has layers — review code before you run it.
latestvk9755x048m7z93c40htkzt5pdn83dn67
License
MIT-0
Free to use, modify, and redistribute. No attribution required.