Back to skill

Security audit

minimax-speech

Security checks for vulnerabilities and agentic risk

Overview

This MiniMax speech skill does what it claims, but users should treat text, voice samples, and remote voice deletion as sensitive actions.

Install only if you intend to use MiniMax's speech APIs. Do not provide private text or voice samples unless you are comfortable sending them to MiniMax, only clone voices with proper consent, choose output paths deliberately, and double-check voice_id values before using deletion.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (12)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding

The skill documents use of environment variables and an external API, but it does not declare any explicit tool scope or permissions boundary. This can lead users or orchestrators to invoke a network-capable skill without clear authorization expectations, increasing the risk of unintended data transmission or over-broad execution in agent environments.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill description explains TTS and voice-cloning features but omits a clear warning that user-provided text and reference audio are sent to the MiniMax API and that generated audio is written to local disk. This is dangerous because users may unknowingly disclose sensitive text, biometric voice data, or create persistent local artifacts without informed consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The documented delete_voice operation is destructive, yet no warning is provided about irreversible deletion or the need to verify the target voice_id. In an agent setting, this increases the chance of accidental or unauthorized deletion of user-created voice assets, especially if the action is triggered from ambiguous prompts or automation.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The document exposes voice-cloning and voice-deletion capabilities without any warning about consent, impersonation risk, privacy handling, or destructive consequences. In a skill that enables direct use of these APIs, missing guardrails can normalize unsafe use of biometric voice data and accidental or unauthorized deletion of voice assets.

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 · references/api.md (reported line 5)May include surrounding context.

md
# API 配置
BASE_URL_CN = "https://api.minimaxi.com/v1"
BASE_URL_INT = "https://api.minimax.io/v1"

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/speech.py (reported line 17)May include surrounding context.

python
# API 配置
BASE_URL_CN = "https://api.minimaxi.com/v1"
BASE_URL_INT = "https://api.minimax.io/v1"

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 · references/api.md (reported line 6)May include surrounding context.

md
# API 配置
BASE_URL_CN = "https://api.minimaxi.com/v1"
BASE_URL_INT = "https://api.minimax.io/v1"


def get_base_url() -> str:

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/speech.py (reported line 18)May include surrounding context.

python
# API 配置
BASE_URL_CN = "https://api.minimaxi.com/v1"
BASE_URL_INT = "https://api.minimax.io/v1"


def get_base_url() -> str:

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The voice cloning function reads a local audio file, base64-encodes it, and uploads it to a third-party API, but the CLI provides no explicit warning that potentially sensitive local content will be transmitted off-host. Audio files can contain biometric voice data and private speech, so silent transmission increases privacy and consent risk, especially in an agent skill context where users may not inspect code.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The delete command performs a remote destructive action immediately based only on a provided voice_id, with no confirmation prompt or safety interlock. In an automation or agent environment, this raises the risk of accidental or unauthorized deletion of remote assets due to bad inputs, prompt injection into higher-level tooling, or operator error.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
71% confidence
Finding

The entire reference is written in Chinese and labels usage for domestic/international users, but it does not indicate that language is optional or that the document is intentionally region-specific. Per the policy rule, forcing a specific language without user opt-in can be a natural-language policy issue unless the locale constraint is clearly documented and justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

User-facing docstrings, comments, and CLI help strings are presented in Chinese throughout the file, while the skill does not offer a language or locale choice. This can violate language/locale policy when a specific language is effectively forced without explicit opt-in or justification.

Content

No source excerpt is available for this finding.