NYC

ai-social-media-content

Fail

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION] (HIGH): The skill instructs the user and agent to execute a remote script using the pattern curl -fsSL https://cli.inference.sh | sh. This is a critical security risk as it downloads and executes unverified code from a non-trusted external source directly into the shell context.
  • [EXTERNAL_DOWNLOADS] (HIGH): The skill references and installs multiple components from inference.sh and npx skills add inference-sh/skills. Since 'inference-sh' is not in the recognized trusted sources list, these represent unverifiable external dependencies.
  • [COMMAND_EXECUTION] (MEDIUM): The skill requests allowed-tools: Bash(infsh *), which grants the AI agent the ability to execute any subcommand of the infsh utility. This provides a broad attack surface if the agent is manipulated into running unintended commands.
  • [PROMPT_INJECTION] (LOW): The skill uses shell variable interpolation (e.g., "$CONCEPT" and "$topic") to build command strings from user-controlled data. This lacks sanitization or boundary markers, creating a surface for indirect prompt injection where a user could provide malformed input to break the JSON structure or execute unintended logic within the infsh tool.
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
Feb 18, 2026, 11:28 PM