Security audit

Withings Family

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Withings integration that retrieves family health data and stores OAuth tokens locally, with privacy precautions users should understand before use.

Install only if each family member whose Withings account is added has agreed to that access. Treat the local token files as sensitive credentials, keep the skill directory private, delete token files or revoke the Withings app when access is no longer needed, and use explicit user IDs and date or result limits when asking for health history.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 94% confidence
Finding: The skill documentation indicates capabilities to read environment variables, read/write local files, invoke shell commands, and make network requests, but it does not declare corresponding permissions. That creates a transparency and governance gap: operators may approve or invoke the skill without understanding that it stores OAuth tokens and accesses sensitive health data from external services.

Vague Triggers

Medium

Confidence: 84% confidence
Finding: The top-level description says the skill should be used for requests about the user's or their family's health measurements, which is broader than clearly Withings-specific requests. In an agent environment, broad triggers can cause unintended invocation on generic health questions, leading to unnecessary retrieval or exposure of highly sensitive family health data.

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The 'When to Use This Skill' section includes generic triggers such as weight, body composition, sleep, and family health progress without requiring a clear Withings context. Because the skill supports multi-user access and persistent tokens, accidental invocation could expose stored health records for the wrong person or without sufficiently explicit user intent.

Missing User Warnings

High

Confidence: 96% confidence
Finding: The skill is designed to retrieve and store sensitive health data for multiple family members, but the description does not prominently warn about the privacy implications of handling medical and behavioral data. In a multi-user setting, absence of explicit privacy guidance increases the risk of unauthorized access, mistaken cross-user retrieval, and insufficient user consent for storing persistent OAuth tokens tied to health records.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: OAuth tokens are stored on disk in a local JSON file, and the skill handles highly sensitive health data. Although the file permission is tightened to 0600 on a best-effort basis, there is no guaranteed secure storage, no encryption at rest, and no user-facing disclosure that sensitive credentials and health-adjacent data are persisted locally.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.