ai-avatar-video
Fail
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- REMOTE_CODE_EXECUTION (HIGH): The skill explicitly instructs the agent or user to execute
curl -fsSL https://cli.inference.sh | sh. This pattern is highly dangerous as it downloads a shell script from an untrusted source and executes it immediately with no integrity verification. An attacker compromising theinference.shdomain could execute arbitrary commands on the underlying host. - EXTERNAL_DOWNLOADS (MEDIUM): The skill promotes the use of
npx skills addto install further dependencies frominference-sh/skills. These external modules are not part of the trusted organization list and could contain malicious logic or further insecure dependencies. - PROMPT_INJECTION (LOW): (Category 8
- Indirect Prompt Injection)
- Ingestion points: Untrusted data enters the skill through several JSON fields:
image_url,audio_url,video_url, andtext(for TTS). - Boundary markers: There are no boundary markers or instructions to the agent to treat these inputs as data rather than potential commands.
- Capability inventory: The skill executes the
infshbinary via the Bash tool, passing these untrusted inputs as arguments. - Sanitization: No evidence of sanitization or validation of the input strings is present in the skill's instructions.
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