NYC

text-to-speech

Fail

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION] (HIGH): The skill documentation includes a 'Quick Start' instruction to execute curl -fsSL https://cli.inference.sh | sh. This is a high-risk pattern that downloads and executes unverified code from the internet directly in the user's shell environment.
  • [EXTERNAL_DOWNLOADS] (HIGH): The skill depends on tools and scripts hosted at inference.sh. As this domain is not part of the defined list of Trusted External Sources, it carries a higher risk for supply chain attacks or unauthorized code execution.
  • [COMMAND_EXECUTION] (MEDIUM): The skill requests allowed-tools: Bash(infsh *). This grants the agent broad permission to execute any sub-command of the infsh tool. If the agent is targeted via indirect prompt injection, it could be coerced into running unauthorized apps or functions within the CLI ecosystem.
  • [DATA_EXFILTRATION] (LOW): The skill documentation suggests running infsh login. While necessary for the service, users should be aware that this process involves the creation and storage of authentication tokens by a third-party utility.
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
Risk Level
HIGH
Analyzed
Feb 18, 2026, 11:31 PM