Drug Interaction Checker

v0.1.0

Check for drug-drug interactions between multiple medications. Trigger when user asks about medication compatibility, "can I take X with Y", drug interaction...

0· 149·2 current·3 all-time
byAIpoch@aipoch-ai
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
medium confidence
Purpose & Capability
Name/description (drug interaction checking) align with included assets: local interaction DB, CYP450 metadata, severity criteria, and a Python script. No unrelated credentials, binaries, or config paths are requested.
Instruction Scope
SKILL.md instructs running the bundled Python script or importing its module and references only the local references/ files. That scope is appropriate, but the runtime will execute code from scripts/main.py — the instructions correctly warn about sandboxing and input validation. Confirm the script doesn’t read arbitrary paths or call external endpoints before running.
Install Mechanism
No install spec or external downloads; dependencies are minimal (requirements.txt lists standard libs). This is low-risk and consistent with an instruction-only skill that includes a local script and data files.
Credentials
The skill requests no environment variables, credentials, or special config paths. All required data is bundled locally, which is proportional to the stated functionality.
Persistence & Privilege
Skill is not always-enabled and does not declare elevated persistence. It does execute local code when invoked but does not request to modify other skills or system-wide configurations.
Assessment
This skill appears internally consistent and self-contained, but it executes a local Python script and uses a bundled database — before installing or running it: (1) quickly scan scripts/main.py for any network calls, subprocess.exec/eval/exec usage, or file reads outside the references/ directory; (2) run it in an isolated environment (container or VM) the first time; (3) verify the interaction data currency and provenance if you will rely on it clinically; and (4) do not treat results as definitive medical advice — confirm with a clinician for patient-specific decisions. If you want higher confidence, paste the full scripts/main.py here (or confirm it makes no external calls) so I can re-check for hidden I/O or exfiltration patterns.

Like a lobster shell, security has layers — review code before you run it.

latestvk97azmyqvrm05yzm0gt799r5yx8362g2

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments