speech-to-text
Fail
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- Remote Code Execution (CRITICAL): The Quick Start section contains the command
curl -fsSL https://cli.inference.sh | sh, which downloads and executes a script from an untrusted third-party domain without verification. - External Downloads (HIGH): The skill requires downloading a CLI tool from
inference.sh, which is not a recognized trusted external source. - Command Execution (MEDIUM): The skill requests permissions for
Bash(infsh *), which allows for arbitrary sub-command execution of the unverified CLI tool. - Prompt Injection (LOW): The skill has a surface for indirect prompt injection. Ingestion points: Untrusted audio and video URLs are processed. Boundary markers: No markers or 'ignore' instructions are present to delimit untrusted data. Capability inventory: The skill can execute commands and access the network via
infsh. Sanitization: No evidence of input validation or sanitization is present.
Recommendations
- HIGH: Downloads and executes remote code from: https://cli.inference.sh - DO NOT USE without thorough review
- AI detected serious security threats
Audit Metadata