zeecu-device-skill

• Outcome: Review
• Latest audit: Jun 25, 2026 · 6:40 AM UTC
• Version: 1.0.2

Overview

The skill does what it claims, but it handles live vehicle location and trip history while silently saving the user's API key to a local plaintext config file.

Install only if you are comfortable granting access to precise vehicle location, trip history, and identifying vehicle details. Prefer using a temporary environment variable over command-line arguments, inspect or delete config.json after use, and rotate the API key if it may have been stored or exposed unintentionally.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (6)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 93% confidence
• Finding: The skill documentation instructs use of environment variables, local config files, and networked query scripts, but the manifest does not explicitly declare permissions for file access, network access, or configuration writes. This creates a transparency and policy gap: a skill that processes API keys and sensitive vehicle telemetry can access more capability than users or reviewers are clearly told about.

Description-Behavior Mismatch

Medium

• Confidence: 94% confidence
• Finding: The script persists the resolved API key into ../config.json on every run, creating local secret storage unrelated to the core read-only vehicle-query function. If that file is readable by other local users, packaged into logs/backups, or accidentally committed, the API key can be reused to access vehicle data and trip history.

Missing User Warnings

Medium

• Confidence: 94% confidence
• Finding: The README explicitly instructs users to pass a live API key on the command line, which can expose the secret through shell history, process listings, terminal logging, CI logs, or shared screenshots. In this skill's context, the key grants access to bound vehicle data and real-time status, so leakage could enable unauthorized access to sensitive mobility information.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The skill handles highly sensitive personal data, including real-time vehicle location, home/work-adjacent addresses, and trip history, yet provides no explicit privacy notice, consent flow, retention statement, or sharing limitations. Exposure of this data could enable stalking, surveillance, profiling, or physical theft, making the context significantly more dangerous than ordinary telemetry.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The document explicitly describes transmitting a live API key and highly sensitive vehicle telemetry, including precise location, address, and trip history, but provides no privacy, storage, masking, or consent guidance. In the context of an agent skill, this can enable exposure of secrets and real-world movement data if implementers log requests, echo examples, or mishandle responses.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The code stores the API key to config.json automatically and without any user-facing warning, confirmation, or opt-in. This increases the chance that users unknowingly leave long-lived credentials on disk, where they may be exposed through backups, multi-user access, artifact collection, or source-control mistakes.

VirusTotal

VirusTotal findings are pending for this skill version.

Static analysis

No suspicious patterns detected.