Back to skill

Security audit

Byted Voice Notify

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly coherent for sending Volcengine voice notifications, but it can place real third-party phone calls from broad prompts without an explicit confirmation step.

Install only if you intend to let the agent use your Volcengine voice-service account to look up voice resources and number pools and send voice calls to supplied phone numbers. Before use, require the agent to confirm the phone number, selected voice resource, number pool, provider, and any cost or privacy impact before every send action.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (9)

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding

声明仅描述“发送语音通知”这一单一能力,但实际代码除发送(SingleBatchAppend)外,还包含两个额外的查询能力:查询语音资源和查询号码池。这些是对外可用的功能,不只是底层实现细节,因此属于未声明能力。另外,代码依赖环境变量中的 AK/SK 凭证,并暴露命令行入口支持 resource、pool、send 三种触发方式;声明中的 triggers 和 permissions 为空,未体现这些实际行为。虽然主要领域仍然是火山引擎语音服务,且发送通知是核心功能之一,但描述未准确覆盖代码的实际能力范围,因此应判定为存在描述-行为不匹配。

Content

No source excerpt is available for this finding.

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 206)May include surrounding context.

md
## 环境变量

使用前需配置火山引擎访问密钥到系统环境变量或 OpenClaw 的 .env 文件:

- **VOLCENGINE_ACCESS_KEY**:火山引擎访问密钥ID
- **VOLCENGINE_SECRET_KEY**:火山引擎秘密访问密钥

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding

The skill requires environment credentials and makes external API calls, but it does not declare an explicit tool scope such as allowed tools or permissions. In an agent setting, missing scope boundaries can cause the model or runtime to invoke shell, network, or secret-bearing capabilities more broadly than users expect, increasing the chance of unintended outbound actions or credential exposure.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The activation condition says to use the skill whenever a user needs to send a voice notification, which is overly broad for a capability that places external voice calls. Ambiguous activation increases the chance that the agent will invoke the skill without sufficiently verifying intent, recipient, or consent.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

Broad trigger phrases like '打语音电话' and regex-style patterns for sending to arbitrary recipients can cause the skill to activate on ordinary conversation or loosely related requests. In this context, mistaken activation can lead to real telephony actions against a phone number via an external provider, creating spam, privacy, and cost risks.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill does not prominently warn that it sends a voice notification to a phone number through an external third-party provider. Without a clear warning and confirmation, users may unknowingly disclose personal data or trigger billable outbound communications.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The create_voice_notify function sends a phone number and related notification parameters over the network to an external voice service. While the script's usage text says it will 'send' a voice notification, there is no explicit privacy or impact warning, confirmation step, or comment/docstring disclosing that personal data is transmitted to a third-party service.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

The manifest describes this skill as being used to send voice notifications to specified phone numbers, but the code also exposes commands to query voice resources and number pools. Those discovery capabilities expand the skill beyond pure notification sending, even though they are related to the same service.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
96% confidence
Finding

The script emits all operational and error messages in Chinese only, including setup instructions and command usage. This imposes a fixed language on users without opt-in or an alternative locale, which matches the language/locale policy violation criteria.

Content

No source excerpt is available for this finding.