AGENT LAB: SKILLS

logo-design-guide

Fail

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION] (CRITICAL): The skill instructs the agent to execute a remote script directly into the shell using curl -fsSL https://cli.inference.sh | sh. This pattern is highly dangerous as it grants the remote server full control over the execution environment.
  • [EXTERNAL_DOWNLOADS] (HIGH): The skill attempts to download and install external agent skills from an untrusted source (inference-sh/skills) using npx skills add. This could lead to the installation of malicious agent extensions.
  • [COMMAND_EXECUTION] (MEDIUM): The skill frequently uses the infsh command-line tool with arbitrary input. While this appears to be the primary purpose of the skill, the reliance on an externally downloaded binary (infsh) that was installed via the insecure curl | sh method escalates the risk of all subsequent command executions.
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
CRITICAL
Analyzed
Feb 19, 2026, 03:41 AM