text-to-speech

Warn

Audited by Gen Agent Trust Hub on Mar 26, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The bash script example provided for batch processing is vulnerable to command injection. Evidence: In the 'Batch Text-to-Speech' section, the script interpolates the $TEXT variable directly into a shell command payload ("content": "... $TEXT"). If the input text contains shell metacharacters such as backticks or command substitution syntax, the shell will execute them during command assembly.
  • [EXTERNAL_DOWNLOADS]: The skill performs network operations to interact with an external speech synthesis service. Evidence: The instructions specify sending user data to https://eachsense-agent.core.eachlabs.run/v1/chat/completions using tools like curl or the openai Python SDK.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 26, 2026, 12:03 PM
Security Audit — agent-trust-hub — text-to-speech