Survival Curve Risk Table

Analyze data with `survival-curve-risk-table` using a reproducible workflow, explicit validation, and structured outputs for review-ready interpretation.

This package appears to be what it claims: a local tool to compute and render 'number at risk' tables for Kaplan–Meier plots. Before installing/using: 1) Only run the script on datasets you trust — the loader supports pandas.read_pickle which will execute code when loading malicious pickle files; prefer CSV/Excel input from trusted sources. 2) Pin dependency versions (requirements.txt has unspecified versions and lists both 'pil' and 'pillow') to avoid supply-chain surprises. 3) Run the suggested smoke-check (python -m py_compile scripts/main.py) and, when possible, test in an isolated environment (container or VM) before using with sensitive clinical data. 4) If you plan to enable autonomous agent invocation, restrict it or review expected input/output flows so the agent can't be prompted to load arbitrary local files without explicit confirmation.

Type: OpenClaw Skill Name: survival-curve-risk-table Version: 1.0.0 The skill is a functional tool for survival analysis but contains security vulnerabilities. The script scripts/main.py uses pd.read_pickle, which allows for arbitrary code execution (RCE) if a malicious file is provided. Additionally, the script lacks explicit path traversal sanitization for its file-handling arguments, and the SKILL.md file contains inconsistent example commands that suggest a lack of thorough validation during the skill's creation.