Back to skill

Security audit

fun-voice-type

Security checks for vulnerabilities and agentic risk

Overview

This skill does what it claims, but users should understand it records microphone audio for cloud speech recognition and can type into any focused app while running.

Install only if you are comfortable granting microphone, Accessibility, and Input Monitoring permissions to the terminal or Python process running this tool. Avoid dictating secrets or regulated information because audio and translated text can be sent to DashScope/Qwen cloud services. Prefer a virtual environment and pinned dependencies before granting macOS privacy permissions.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:25
Finding

Unpinned Third-Party Dependencies Create a Supply-Chain Risk

Content
View full analysis
Remediation
View remediation
pynput== pyaudio== pystray== Pillow== ``` 2. Generate and require cryptographic hashes, for example with `pip-compile --generate-hashes`, and install using: ```bash python -m pip install --require-hashes -r requirements.txt ``` 3. Pin or otherwise document a reviewed Homebrew `portaudio` version and verify its source and checksum where the package manager permits. 4. Install dependencies inside a dedicated virtual environment rather than the user's global Python environment. 5. Commit the lockfile or hash-locked requirements file to the project and review dependency updates before changing it. 6. Add automated dependency vulnerability and provenance checks to the release process. 7. Clearly identify which process requires macOS privacy permissions and advise users not to grant those permissions until dependency installation and verification are complete. ]]>
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (6)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
84% confidence
Finding

The skill requests or documents use of environment variables for a cloud API key but does not declare any explicit tool scope or permissions metadata. This creates a transparency and governance gap: reviewers and users cannot easily see that the skill depends on secret material and external service access, increasing the chance of overbroad execution in agent environments.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The activation examples include broad phrases such as '帮我记录这段话' and '我不方便打字', which could match ordinary conversation and trigger the skill unintentionally. In this skill's context, accidental activation is more dangerous because it can start microphone capture and inject transcribed text into any focused input field, potentially exposing sensitive speech or causing unintended actions.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
65% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 39)May include surrounding context.

安装与运行

运行脚本,fun-voice-type将显示为Mac菜单栏右上角的小图标:

bash
nohup python fun-voice-type.py > /dev/null 2>&1

此时长按右Option即可实现语音输入功能。

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This code captures live microphone audio and streams it to DashScope's remote ASR service as soon as the hotkey is pressed, but it provides no clear runtime consent flow, privacy notice, or indication that speech leaves the local device. Because microphone data can contain sensitive conversations, credentials, or personal information, silent transmission to a third party creates a real privacy and data-handling risk.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

When translation is enabled, recognized text is embedded into a prompt and sent to a remote LLM service without a user-facing warning or per-use confirmation. Transcribed text may include sensitive messages, secrets, or regulated data, so forwarding it to a second cloud service materially expands exposure beyond the original speech-recognition function.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

All user-facing natural-language instructions and activation examples are presented only in Chinese, and the file does not state that the skill is intentionally limited to Chinese-speaking users or a China-specific deployment. Under the policy, forcing a specific language without user opt-in or justification is a natural-language locale violation.

Content

No source excerpt is available for this finding.