book-cover-design
Fail
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- Remote Code Execution (CRITICAL): The skill provides a command
curl -fsSL https://cli.inference.sh | shwhich pipes remote content directly into a shell. This is a classic attack vector that allows an untrusted third party to execute arbitrary code on the host system without verification. The domaininference.shis not on the list of trusted external sources. - External Downloads (HIGH): The skill uses
npx skills add inference-sh/skills@...to fetch and integrate additional logic. Since theinference-shorganization is not a trusted provider, these dependencies are unverifiable and could contain malicious payloads. - Command Execution (HIGH): The skill relies extensively on the
infshCLI tool. Given its installation method (piped shell script) and untrusted origin, the tool's execution of remote 'apps' represents a significant security risk where the agent's capabilities are handed over to an unvetted binary.
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