Protocol Deviation Classifier
v1.0.1Determine whether an incident in a clinical trial is a "major deviation.
⭐ 0· 28·0 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
medium confidencePurpose & Capability
Name/description match the packaged artifact: a Python classifier (scripts/main.py) that uses keyword patterns and risk heuristics to classify deviations. Required env, binaries, and config paths are empty and appropriate for a standalone classifier.
Instruction Scope
SKILL.md instructs the agent to validate inputs and run the packaged script only (python -m py_compile and python scripts/main.py). The instructions do not ask the agent to read unrelated system files, harvest credentials, or post data to third‑party endpoints.
Install Mechanism
No install spec is provided (instruction-only skill with included script). Files are local to the package; there is no download-from-URL or archive extraction step in the manifest.
Credentials
The skill requires no environment variables or external credentials. The declared requirements (Python 3.8+, minimal stdlib) are proportional to the described functionality.
Persistence & Privilege
always is false (no forced inclusion), and there is no indication the skill attempts to modify other skills or system-wide agent settings. Autonomous invocation is allowed by default but not by itself a problem.
Assessment
This package appears coherent for classifying protocol deviations, but before running it on real clinical data: 1) perform a quick code review of scripts/main.py to ensure there are no network calls or any code that transmits data (look for imports/use of requests, urllib, socket, http.client, ftplib, paramiko, subprocess/os.system, or hardcoded URLs/IPs); 2) verify logging/storage behavior (ensure PHI is not written to shared locations or uploaded); 3) run the included smoke checks (python -m py_compile scripts/main.py and python scripts/main.py --help) and test on synthetic/deidentified examples; 4) confirm the regulatory rationale and recommended actions meet your internal SOPs before relying on outputs for compliance decisions. If you want, I can scan the full scripts/main.py content for network I/O, file writes, or subprocess usage — provide the remainder of the file and I will review it line-by-line.Like a lobster shell, security has layers — review code before you run it.
latestvk974phtwnek4e1wak34rcs7mx1843dt7
License
MIT-0
Free to use, modify, and redistribute. No attribution required.