ai-product-photography
Fail
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- Unverifiable Dependencies & Remote Code Execution (HIGH): The skill instructs the user to execute
curl -fsSL https://cli.inference.sh | shto install a CLI. This pattern executes unverified remote code. Asinference.shis not a trusted source, this is a high-risk vector. - External Downloads (HIGH): The skill downloads and runs scripts from an untrusted external domain.
- Command Execution (LOW): The skill uses the
infshcommand-line tool via the Bash provider. - Indirect Prompt Injection (LOW): The skill ingests user input for prompts and interpolates them into JSON inputs for tool calls without sanitization or boundary markers. Ingestion point: Prompt fields in
infsh app runcommands inSKILL.md. Capability inventory: Bash tool execution. Sanitization: None identified.
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