Unstructured Medical Text Miner
v1.0.0Mine unstructured clinical text from MIMIC-IV to extract diagnostic logic.
⭐ 0· 33·0 current·0 all-time
byAIpoch@aipoch-ai
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description match the included code and SKILL.md. The package contains Python scripts that implement regex- and spaCy-based entity/relation extraction and timeline logic consistent with extracting diagnostic logic from MIMIC‑IV text. Declared NLP dependencies (spacy, scispacy, negspacy, pandas, etc.) are proportionate to the task.
Instruction Scope
SKILL.md instructs local execution of scripts (python -m py_compile, python scripts/main.py) and to provide MIMIC‑IV note files; it does not instruct reading unrelated system files or contacting external endpoints. Caution: the skill expects you to edit/configure parameters (in-file CONFIG) and will read/process sensitive clinical text. Review export/save logic (e.g., export_to_json) to ensure no unintended disclosure.
Install Mechanism
There is no automated install spec; the repo includes requirements.txt listing standard Python/NLP packages. No arbitrary remote download URLs or archive extraction were found. Note that scispacy/spacy models (e.g., en_core_sci_lg) may require additional model downloads and substantial disk/IO/network when installed.
Credentials
The skill requests no environment variables, credentials, or config paths. This is proportionate for a local data‑processing NLP tool. Important: although it doesn't request secrets, it is designed to process highly sensitive MIMIC‑IV data (PHI), so data governance and access controls are required before use.
Persistence & Privilege
always is false and there is no indication the skill modifies other skills or system-wide agent settings. The default ability for an agent to invoke skills autonomously is present but not combined with other concerning flags.
Assessment
This skill appears coherent for local analysis of MIMIC‑IV notes. Before installing/running: (1) confirm you have lawful access to MIMIC‑IV and follow your institution's PHI handling rules; (2) review scripts/main.py and any export functions to ensure outputs are written to secure locations and not transmitted externally; (3) install recommended Python packages and the required spaCy/scispaCy models in an isolated environment (virtualenv or container) and be prepared for large model downloads; (4) test on synthetic or fully deidentified data first; (5) if you need to run in a shared environment, restrict file permissions and logs so patient data cannot leak. If you want, I can list specific lines in scripts/main.py that write files or otherwise persist output so you can inspect them more closely.Like a lobster shell, security has layers — review code before you run it.
latestvk976yd3c088dw00frb7cbx3ny1840d39
License
MIT-0
Free to use, modify, and redistribute. No attribution required.