Back to skill

Security audit

Xiaomi MiMo-V2-TTS

Security checks for vulnerabilities and agentic risk

Overview

The skill’s text-to-speech purpose is coherent, but it automatically installs an unpinned Python package at runtime, which users should review before installing.

Review this skill before installing. It appears to perform the advertised Xiaomi MiMo TTS task, but install requests through your normal trusted dependency process first and avoid running the script in an environment where it can auto-install packages. Use MIMO_API_KEY rather than --api-key when possible, and do not send confidential, regulated, or secret text unless you are comfortable sharing it with the Xiaomi MiMo API and storing the generated audio file locally.

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
scripts/mimo_tts.py:25
Finding

Automatic Installation of an Unpinned Runtime Dependency

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Note
Location
scripts/mimo_tts.py:166
Finding

API Key Exposure Through Command-Line Arguments

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (8)

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

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

Content

Scanner excerpt · scripts/mimo_tts.py (reported line 41)May include surrounding context.

python
def get_api_key(cli_key: str = None) -> str:
    """Get API key from CLI arg or MIMO_API_KEY env var."""
    if cli_key:
        return cli_key
    key = os.environ.get("MIMO_API_KEY")

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill documentation instructs users to send arbitrary input text to Xiaomi's external API and save the generated audio to a local file, but it does not clearly warn about the privacy and data-handling implications. This can lead users or downstream agents to transmit sensitive text off-device and persist potentially sensitive spoken content locally without informed consent or clear disclosure.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
88% confidence
Finding

The skill is explicitly designed to communicate with an external service at api.xiaomimimo.com, which means user-provided text leaves the local environment and is exposed to third-party processing. In a TTS skill this behavior is expected, but it still creates confidentiality and compliance risk if users provide sensitive content without a clear disclosure or policy controls.

Content

Scanner excerpt · SKILL.md (reported line 12)May include surrounding context.

md
## API Details
- **Platform**: https://platform.xiaomimimo.com
- **Base URL**: `https://api.xiaomimimo.com/v1`
- **Endpoint**: `/v1/chat/completions` (NOT `/audio/speech`)
- **Model**: `mimo-v2-tts`
- **Auth**: Bearer Token via `MIMO_API_KEY` env var

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/mimo_tts.py (reported line 5)May include surrounding context.

python
"""
Xiaomi MiMo-V2-TTS - Text-to-Speech synthesis script.

API: https://api.xiaomimimo.com/v1/chat/completions
Model: mimo-v2-tts

MiMo TTS uses Chat Completions format with special requirements:

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/mimo_tts.py (reported line 35)May include surrounding context.

python
"""
Xiaomi MiMo-V2-TTS - Text-to-Speech synthesis script.

API: https://api.xiaomimimo.com/v1/chat/completions
Model: mimo-v2-tts

MiMo TTS uses Chat Completions format with special requirements:

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

Auto-installing a package from inside a TTS script is unnecessary and dangerous because it triggers network access and package execution as a side effect of running the tool. If the package source, environment, or dependency chain is compromised, the script can facilitate unintended code execution.

Content

No source excerpt is available for this finding.

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
96% confidence
Finding

The script automatically executes pip via subprocess when requests is missing, causing code installation and execution at runtime without user confirmation. This expands the trust boundary to package indexes and the local Python environment, creating supply-chain and unexpected code-execution risk that is unrelated to core TTS functionality.

Content

Scanner excerpt · scripts/mimo_tts.py (reported line 31)May include surrounding context.

python
import requests
except ImportError:
    import subprocess
    subprocess.check_call([sys.executable, "-m", "pip", "install", "requests", "-q"])
    import requests

External Transmission

Medium
Category
Data Exfiltration
Confidence
80% 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/mimo_tts.py (reported line 110)May include surrounding context.

python
print(f"  Format: {response_format}")

    try:
        resp = requests.post(url, headers=headers, json=payload, timeout=120)

        if resp.status_code == 200:
            data = resp.json()