Back to skill

Security audit

iwown device gnss tracker

Security checks for vulnerabilities and agentic risk

Overview

This GNSS location skill is purpose-aligned, but it can query sensitive device location by arbitrary IMEI through a third-party service without clear consent or ownership checks.

Review before installing. This skill is for device location lookup, so only use it where you have clear authority to locate the device and its user. The publisher should add explicit disclosure that IMEIs are sent to search.iwown.com, require user confirmation and ownership/authorization checks, and limit output to the minimum location fields needed.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Findings (2)

T05 · Unauthorized Access and Privilege Escalation

Error
Location
scripts/query_gnss.py:6
Finding

Device Location Lookup Lacks Caller Authorization and Ownership Verification

Content
View full analysis
") sys.exit(1) imei = sys.argv[1] record_time, address, raw_data = query_location(imei) ``` ### Technical Analysis The script accepts an arbitrary IMEI from the command line and requests the device's latest GNSS record. It does not authenticate the caller, verify that the caller owns or manages the referenced device, obtain the tracked person's consent, or supply an authorization credential to the remote service. An IMEI is an identifier, not an authentication secret. Treating possession of an IMEI as sufficient authority to obtain a device's location creates an insecure direct object reference pattern. The workflow documented in `SKILL.md:15-18` also allows an IMEI to be extracted from a QR code and immediately used for the lookup, without an intervening authorization decision. ### Attack Path 1. An attacker obtains a target device's IMEI from a QR code, exposed record, photograph, support interaction, or another source. 2. The attacker invokes `python scripts/query_gnss.py TARGET_IMEI`, or causes the Skill to process a QR code containing that IMEI. 3. The script sends the identifier directly to the GNSS endpoint as the `deviceid` query parameter. 4. If the endpoint accepts the request, the script retrieves and displays the device's latest address and record time. 5. The attacker uses this information to infer the target's current o ...[truncated 492 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/query_gnss.py:31
Finding

Complete Raw GNSS Record Is Disclosed in Script Output

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The skill claims it can recognize QR codes from user-supplied images, but the included code only accepts an IMEI via command line and performs a remote lookup. This mismatch can mislead reviewers and users about what data is processed, what tooling is needed, and what safeguards should apply, which is especially risky in a location-tracking skill handling sensitive identifiers and location records.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding

The skill performs an external network request to a third-party GNSS endpoint but does not declare any tool scope or permissions that would make this capability explicit. This weakens reviewability and consent controls, especially because the queried value is an IMEI and the response contains precise location data, both of which are sensitive.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill sends a device IMEI to an external API and returns device location information, but it does not warn the user that sensitive identifiers and location data are being transmitted off-platform. Because IMEI and geolocation are highly sensitive, lack of disclosure undermines informed consent and increases privacy and compliance risk.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The script sends a device identifier (IMEI) to an external service and returns precise location-related data without any visible consent check, authorization enforcement, or disclosure to the user about the outbound transmission. In the context of a GNSS-tracking skill, this can expose sensitive device location and tracking information to unauthorized parties if the skill is invoked on arbitrary identifiers.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

User-visible output such as the error and usage messages is only provided in Chinese, which enforces a specific language without user opt-in. The file does not indicate that this is a China-specific or otherwise justified locale-restricted tool.

Content

No source excerpt is available for this finding.