Back to skill

Security audit

minimax_ttsvoicereponse_feishu

Security checks for vulnerabilities and agentic risk

Overview

This voice-reply skill has a coherent purpose, but it needs review because it sends user text to MiniMax and its script can overwrite or delete writable files if invoked with an unsafe output path.

Install only if you are comfortable sending voice-reply text to MiniMax. Avoid sensitive or confidential content, use explicit trigger phrases or confirmation before sending, and fix or restrict the output path handling so generated audio stays inside the OpenClaw workspace.

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/voice_reply.py:69
Finding

Caller-Controlled Output Path Enables Arbitrary File Overwrite and Deletion

Content
View full analysis
2 else None result_path = tts_to_ogg(text, output) ``` ### Technical Analysis The second command-line argument is accepted as an output path without validation or confinement to an approved workspace. The script then: 1. Creates parent directories for that path. 2. Derives a predictable MP3 path using string replacement. 3. Opens the derived path using `wb`, truncating an existing file. 4. invokes FFmpeg with `-y`, permitting unconditional overwrite of the requested output. 5. Deletes the derived MP3 path after conversion. Neither the output nor temporary path is canonicalized and checked against an authorized directory. The implementation also does not re ...[truncated 2318 chars]
Remediation
View remediation
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
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
Findings (12)

Tainted flow: 'api_key' from os.environ.get (line 46, credential/environment) → requests.post (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.

Content

Scanner excerpt · scripts/voice_reply.py (reported line 88)May include surrounding context.

python
}
    
    try:
        resp = requests.post(
            API_URL,
            headers={
                "Authorization": f"Bearer {api_key}",

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The description explains the text-to-speech workflow but does not clearly warn users at the point of use that their text is sent to MiniMax, a third-party external service. This creates a privacy and data-handling risk, especially if users dictate sensitive content assuming the conversion is local or only within Feishu.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrases are broad and include ordinary conversational language such as '说给我听' and '语音回复', which can cause the skill to activate unintentionally during normal chat. In this skill, accidental activation is more concerning because it can transmit user-provided text to an external TTS provider and send messages on Feishu without a sufficiently explicit consent boundary.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The natural-language comments and usage instructions are entirely in Chinese, which imposes a specific language on users without any visible opt-in or alternative. Under the policy, language constraints should either offer user choice or be clearly justified as region-specific.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The README explains converting user-provided text into voice and sending it via Feishu, but it does not clearly disclose that the text content will be sent to an external MiniMax TTS provider. This creates a real privacy and data-handling risk because users may provide sensitive content without informed consent, especially in an agent context where prompts or messages can contain confidential information.

Content

No source excerpt is available for this finding.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
70% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · references/README.md (reported line 17)May include surrounding context.

bash
# Ubuntu/Debian
sudo apt install ffmpeg

# macOS
brew install ffmpeg

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The natural-language instructions and activation example are presented only in Chinese, including the stated trigger phrase "语音回复". Under the stated policy, forcing a specific language without user opt-in or documented justification is a locale/language policy violation.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/voice_reply.py (reported line 28)May include surrounding context.

python
# ========== 配置区 ==========
# MiniMax TTS API 端点
API_URL = "https://api.minimaxi.com/v1/t2a_v2"

# 默认输出路径
DEFAULT_OUTPUT = os.path.join(os.path.expanduser("~"), ".openclaw", "workspace", "voice_reply.ogg")

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill sends arbitrary user-provided text to a third-party TTS API, but the script does not provide a clear runtime disclosure or consent step before transmitting potentially sensitive content off-device. In an assistant/skill context, users may reasonably assume local processing, so this can cause unintended exposure of private prompts, personal data, or confidential content.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
92% confidence
Finding

This code performs external transmission of user-supplied text to a remote TTS provider. In the context of a voice-reply skill, external transmission is functionally expected, but it still creates a genuine privacy and data-exposure risk if users are not clearly informed that their content leaves the local environment.

Content

Scanner excerpt · scripts/voice_reply.py (reported line 88)May include surrounding context.

python
}
    
    try:
        resp = requests.post(
            API_URL,
            headers={
                "Authorization": f"Bearer {api_key}",

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/voice_reply.py (reported line 113)May include surrounding context.

python
f.write(audio_bytes)
    
    # Step 2: 转 OGG (Opus) - 先重采样到48000Hz(opus支持)
    ffmpeg_result = subprocess.run([
        "ffmpeg", "-i", mp3_path,
        "-c:a", "libopus", "-b:a", "128k",
        "-ar", "48000",

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
96% confidence
Finding

Natural-language policy violations apply to all file types, including markdown. This README presents the skill description and instructions in Chinese only, which can constitute a forced language choice when no opt-in, alternative language, or documented region-specific justification is provided.

Content

No source excerpt is available for this finding.