ai-avatar-video
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: CRITICALREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- REMOTE_CODE_EXECUTION (CRITICAL): The skill instructs the user/agent to execute
curl -fsSL https://cli.inference.sh | sh. This pattern is highly dangerous as it executes unverified code from an external, untrusted source with the permissions of the current user. - EXTERNAL_DOWNLOADS (HIGH): The skill relies on an external binary and installation script from
inference.sh, which is not a trusted source according to the established security policy. - COMMAND_EXECUTION (HIGH): The skill uses the
Bashtool to executeinfshcommands. Since the source ofinfshis untrusted, these commands represent a high risk of malicious side effects. - INDIRECT_PROMPT_INJECTION (HIGH): The skill processes untrusted external data in the form of
image_url,audio_url, andvideo_url. There is no evidence of sanitization or boundary markers (Category 8). An attacker could provide a URL that, when processed by theinfshCLI, triggers unexpected behavior or exfiltrates local information via the CLI's network capabilities.
Recommendations
- CRITICAL: Downloads and executes remote code from untrusted source(s): https://cli.inference.sh - DO NOT USE
- AI detected serious security threats
Audit Metadata