Skills
skills/inference-sh-3/skills/linkedin-content/Gen Agent Trust Hub

linkedin-content

Fail

Audited by Gen Agent Trust Hub on Mar 25, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill documentation includes a command that downloads and executes a shell script from a remote URL to install the vendor's CLI tool.
  • Evidence: curl -fsSL https://cli.inference.sh | sh in SKILL.md.
  • Context: The download targets a domain associated with the skill's author.
  • [COMMAND_EXECUTION]: The skill utilizes the infsh command-line utility to perform various tasks, including logging in, executing research agents, and generating visual content.
  • Evidence: Multiple instances of infsh app run and infsh login in SKILL.md.
  • [PROMPT_INJECTION]: The skill is configured to interpolate user-provided content directly into shell command arguments, which serves as a surface for indirect prompt injection.
  • Ingestion points: The --input JSON payload in various infsh app run commands within SKILL.md.
  • Boundary markers: Input data is structured within JSON strings.
  • Capability inventory: The infsh tool can perform network operations and interact with external APIs.
  • Sanitization: No explicit sanitization or input validation logic is described in the provided skill instructions.
Recommendations
  • HIGH: Downloads and executes remote code from: https://cli.inference.sh - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Mar 25, 2026, 01:02 AM