ClawHub
Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Phenosnap Phenotype Extractor

v0.1.1

Extract clinical phenotypes and medication entities from user-provided text using PhenoSnap, producing a timestamped JSON output.

⭐ 0· 562·0 current·0 all-time
byKai Wang@kaichop
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Benign
medium confidence
✓
Purpose & Capability
Name/description (phenotype + medication extraction) match the declared needs: python3 to run PhenoSnap, a way to fetch PhenoSnap (git/curl/PowerShell), and an HPO OBO file (HPO_OBO_PATH) for ontology lookup. These requirements are proportionate to the stated purpose.
ℹ
Instruction Scope
SKILL.md stays on task (redact user PHI, write inputs/outputs locally, run PhenoSnap). It explicitly clones or downloads PhenoSnap and may auto-install Python dependencies and run PhenoSnap scripts — i.e., it instructs the agent to fetch and execute third‑party code. Redaction is required but implemented as heuristics (may miss identifiers). The skill writes timestamped files to disk under baseDir.
ℹ
Install Mechanism
No formal install spec; runtime instructions fetch code from GitHub (https://github.com/WGLab/PhenoSnap) via git or downloadable zip — a common and traceable source. There is fallback to pip/get-pip.py and running pip installs, which increases attack surface. URLs used are GitHub releases/branches (not shorteners or unknown personal servers), so risk is moderate but expected for this type of skill.
ℹ
Credentials
The only required env var is HPO_OBO_PATH (a path to an ontology file), which is appropriate. It is odd that this path is declared as the 'primary credential' (primaryEnv) even though it isn't a secret; this is a labeling quirk but not materially harmful. No API keys or unrelated secrets are requested.
✓
Persistence & Privilege
always:false (not force-included). The skill writes artifacts under a declared baseDir and creates local directories; it does not request system-wide config changes or other skills' credentials. Autonomous invocation is enabled (platform default) but not a special privilege here.
Assessment
This skill will download PhenoSnap from GitHub and run its Python code locally and may auto-install Python packages. Before installing/use: (1) supply hp.obo locally and set HPO_OBO_PATH; (2) review the upstream WGLab/PhenoSnap repository code to ensure you trust it; (3) run the skill in a restricted environment or virtualenv (avoid running as admin); (4) verify the redaction step on any sensitive clinical text — heuristics can miss identifiers; (5) be aware the skill will write timestamped input/output files under the chosen baseDir; (6) if you are uncomfortable with automatic pip/get-pip.py network installs or executing third‑party scripts, do not install or audit the repository first. The 'primaryEnv' label for HPO_OBO_PATH is unusual but not itself a secret-exfiltration signal.

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

latestvk97crhmg3t8ydry4vsm4mmq49n81whr5

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Runtime requirements

Binspython3
Any bingit, curl, powershell
EnvHPO_OBO_PATH
Primary envHPO_OBO_PATH

Comments