landing-page-design

Fail

Audited by Gen Agent Trust Hub on Mar 18, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill facilitates remote code execution by instructing the user to run 'curl -fsSL https://cli.inference.sh | sh'. This pattern allows a third-party server to execute arbitrary commands on the local system without prior inspection or verification.
  • [EXTERNAL_DOWNLOADS]: The skill downloads the 'infsh' CLI binary and uses 'npx' to dynamically add further skills from the 'inference-sh' repository. These external components are fetched and executed at runtime, bypassing standard package verification workflows.
  • [COMMAND_EXECUTION]: The skill uses the 'Bash' tool to execute the 'infsh' CLI, which manages interactions with external AI providers and research assistants, performing operations based on the skill's instructions.
  • [PROMPT_INJECTION]: The skill processes untrusted data from the 'tavily/search-assistant' tool to research competitor landing pages. This ingestion point lacks explicit boundary markers or sanitization, creating a surface for indirect prompt injection if external search results contain malicious instructions designed to influence the agent's behavior.
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
HIGH
Analyzed
Mar 18, 2026, 01:53 AM