Back to skill

Security audit

TG Cam App

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed camera-control skill that uses user-provided camera-service credentials to query devices, events, images, status, battery, and snapshots.

Install only if you intend to let this skill access your bound camera devices through the webcamapp.cc service. Be explicit when asking for snapshots, because a snapshot actively captures a new camera image, and treat stored TIVS credentials in openclaw.json as sensitive.

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The top-level description does not clearly warn that the snapshot feature actively instructs the device to capture a new image rather than merely retrieving an existing frame. This is dangerous because users may unknowingly trigger surveillance in a private environment, and the skill itself later notes the action incurs device-side capture, delay, and cost.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The trigger examples are broad everyday phrases such as asking what happened today or checking whether a device is online, which can cause the skill to be invoked in situations where the user did not clearly intend a privacy-sensitive camera action. In this context, unintended invocation is more dangerous because the skill can access device lists, event history, and trigger fresh snapshots of cameras inside private spaces.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The description and operational instructions are presented in Chinese, which may impose a language preference on users or operators without explicit opt-in. Under the language/locale policy, a fixed language should either be justified as region-specific or accompanied by a user choice.

Content

No source excerpt is available for this finding.