ClawHub
Back to skill

Security audit

Azure语音转写免费版

Security checks for vulnerabilities and agentic risk

Overview

The skill is a plausible Azure transcription helper, but it under-discloses sensitive audio exposure and uses overbroad triggers and public Blob guidance that users should review carefully.

Install only if you are comfortable using Azure credentials and sending recordings to Azure for transcription. Do not make confidential audio publicly accessible; use private storage with short-lived, read-only access and avoid logging full Blob or SAS URLs. Treat the translation/localization trigger as incorrect unless the publisher narrows it to audio transcription.

A.I.G

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:167
Finding

Unpinned Third-Party Package Installation

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:235
Finding

Public Blob Access Recommended for Potentially Sensitive Audio

Content
View full analysis
Remediation
View remediation

SkillSpector

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 Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (7)

Vague Triggers

High
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger says to use the skill when '需要文本翻译、多语言转换、本地化处理时使用', which describes broad translation/localization needs rather than Azure audio transcription. This ambiguity could cause the skill to be invoked for many unrelated everyday language tasks outside its intended transcription function.

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 204)May include surrounding context.

环境变量配置

bash
# .env 文件
TRANSCRIPTION_ENDPOINT=https://myresource.cognitiveservices.azure.com
TRANSCRIPTION_KEY=your_subscription_key_here

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The description states '支持中文交互' in a way that presents Chinese interaction as the expected mode, while the rest of the skill also hard-codes Chinese locale examples. There is no explicit user opt-in or language-choice statement, which can violate language/locale policy when a skill appears to force a specific interaction language.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill instructs users to submit Blob-hosted audio URLs and use Azure credentials, but it does not clearly foreground that audio content will be transmitted to an external cloud provider for processing. In a skill context, this can lead to inadvertent disclosure of sensitive recordings, especially for meetings, podcasts-in-draft, or internal media, because users may not realize the privacy and compliance implications of cloud transcription.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The code examples throughout the file implement Azure transcription jobs and output processing, but this section advertises broad operation modes such as create/query/modify/delete, import/reset, and export/save/convert through abstract parameters. Those claims materially overstate the behavior of the skill and do not reflect the concrete transcription workflow shown elsewhere in the file.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest description and main overview consistently position the skill as a batch audio transcription tool that converts speech to text with timestamps. However, the trigger section says to use it for text translation, multilingual conversion, and localization processing, which are different capabilities not implemented or otherwise supported by the documented code examples.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

The description emphasizes easy out-of-the-box use for personal users without complex configuration, but the rest of the file requires Azure Cognitive Services credentials, environment variable setup, and publicly accessible Blob URLs. While these are valid implementation requirements for Azure transcription, they partially conflict with the claim of minimal setup and direct usability.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.