ClawHub
Back to skill

Security audit

Elevenlabs Integration with Openclaw

Security checks across malware telemetry and agentic risk

Overview

ClawVox appears purpose-built for ElevenLabs voice work, but it needs review because it handles sensitive voice data with weak consent/privacy disclosure and can expose the API key in debug or test usage.

Install only if you are comfortable sending selected text, audio, video, and voice samples to ElevenLabs. Use voice cloning only with explicit permission from the speaker, avoid confidential or regulated recordings unless authorized, do not run transcribe.sh with DEBUG enabled in shared logs, and prefer protected environment/config storage over passing the API key on the command line.

SkillSpector

By NVIDIA
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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill documentation clearly instructs users to run local shell scripts and requires external binaries, yet the skill metadata does not declare corresponding permissions. This creates a transparency and policy-enforcement gap: a host system may under-classify the skill's capabilities, causing users or tooling to trust and install a shell-capable skill without appropriate review.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The README promotes voice cloning and multilingual dubbing of audio/video without warning that uploaded media and cloned voices may contain sensitive biometric data, copyrighted content, or third-party speech requiring consent. In this skill context, the omission is more dangerous because the advertised features are specifically designed to process and replicate human voices, which creates meaningful privacy, impersonation, and unauthorized data-transfer risk if users assume the operation is local or unrestricted.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The voice-cloning section encourages uploading voice samples and creating cloned voices without any warning about consent, impersonation risk, or the biometric sensitivity of voice data. Because voiceprints can be used for fraud, deception, or unauthorized identity replication, omission of consent and privacy guidance materially increases the chance of harmful or unlawful use.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The transcription, dubbing, and voice-isolation features instruct users to submit audio files to ElevenLabs but do not clearly warn that potentially sensitive recordings are transmitted to a third-party service for processing. Users may unknowingly upload private conversations, regulated data, or confidential media, creating privacy, compliance, and data-handling risks.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The help text advertises voice cloning from audio samples without any notice about consent, authorization, or privacy risks. In a voice-cloning skill, omission of these safeguards can facilitate impersonation, non-consensual biometric use, and misuse of third-party audio, making the context materially more dangerous.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill documents transcription of user-provided audio but does not disclose that audio may be transmitted to ElevenLabs for external processing. This can lead users to expose sensitive recordings without informed consent, especially in a tool explicitly designed to handle audio content.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The script uploads the user-supplied audio/video file to ElevenLabs' external API, but it does not provide an explicit warning or consent checkpoint at the point of transmission. In a voice-processing skill, uploaded media may contain sensitive speech, biometric voice data, or private conversations, so silent transmission creates a real privacy and compliance risk even if it is the intended functionality.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The script sends user-supplied text and selected voice settings to the ElevenLabs external service, but it does not provide an explicit privacy or data-egress warning at the point of use. In a voice-generation skill, this behavior is expected, but it is still a real privacy/security concern because sensitive prompts or file contents passed via --input may leave the local system without clear user acknowledgement.

Missing User Warnings

Medium
Confidence
99% confidence
Finding
When DEBUG is enabled, the script logs the fully expanded curl command, including the xi-api-key header. This exposes the ElevenLabs API key to terminal history, logs, CI job output, or other observers, enabling unauthorized use of the account and any associated billing or data access.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The script explicitly documents and accepts an API key as a positional command-line argument, then exports it into the environment. Passing secrets on the command line can expose them through shell history, process listings, CI logs, or audit tooling, which creates a real credential leakage risk.

External Transmission

Medium
Category
Data Exfiltration
Content
START_TIME=$(date +%s)

TEMP_OUTPUT="${OUTPUT}.tmp.$$"
HTTP_CODE=$(curl -s -w "%{http_code}" -o "$TEMP_OUTPUT" \
    -X POST "https://api.elevenlabs.io/v1/text-to-speech/$VOICE_ID" \
    -H "xi-api-key: $API_KEY" \
    -H "Content-Type: application/json" \
Confidence
96% confidence
Finding
curl -s -w "%{http_code}" -o "$TEMP_OUTPUT" \ -X POST "https://api.elevenlabs.io/v1/text-to-speech/$VOICE_ID" \ -H "xi-api-key: $API_KEY" \ -H "Content-Type: application/json" \ -d

External Transmission

Medium
Category
Data Exfiltration
Content
log_info "Fetching voice preview..."
        
        RESPONSE=$(curl -s "https://api.elevenlabs.io/v1/voices/$VOICE_ID" \
            -H "xi-api-key: $API_KEY") || {
            log_error "Failed to connect to ElevenLabs API"
            exit 1
Confidence
84% confidence
Finding
https://api.elevenlabs.io/

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.