Security audit

Voxrn

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed Voxrn telephony integration that can place real calls and texts, so it needs careful use but does not show hidden or deceptive behavior.

Install only in trusted OpenClaw workspaces and bridged chats. Store the Voxrn API key in the secrets manager, verify recipients and message bodies before sending texts or calls, expect possible charges, and avoid using alternate VOXRN_BASE_URL values unless you trust that endpoint.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (2)

Missing User Warnings

Medium

Confidence: 86% confidence
Finding: The README advertises tools for placing calls, sending SMS, and reading balances, but it does not warn users that these actions trigger real external communications, may incur telephony/SMS charges, and may expose contact data or message contents to third parties. In an agent skill that bridges chat platforms to the PSTN, lack of explicit warnings increases the risk of accidental misuse, privacy violations, and unexpected costs.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The SMS example sends message content to an external telephony service immediately after contact lookup, without showing an explicit confirmation step or warning that message contents will be transmitted off-platform. In a messaging skill that bridges chats to the PSTN, this increases the risk of unintended disclosure, misdelivery, or acting on prompt-injected/ambiguous requests because SMS is an irreversible external action.

Static analysis

No suspicious patterns detected.