Skills
▥AGENT LAB: SKILLS
skills/inference-sh-3/skills/linkedin-content/Gen Agent Trust Hub

linkedin-content

Fail

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION] (HIGH): The skill contains instructions to execute a remote script directly in the shell using curl -fsSL https://cli.inference.sh | sh. This pattern is extremely dangerous as it grants the remote server full control over the local environment.
  • [EXTERNAL_DOWNLOADS] (HIGH): The skill relies on software from inference.sh, which is not among the defined trusted organizations or repositories. This introduces supply-chain risk.
  • [COMMAND_EXECUTION] (HIGH): The skill is configured with allowed-tools: Bash(infsh *), meaning the agent is explicitly authorized to execute the binary installed via the untrusted remote script.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The skill uses external tools like tavily/search-assistant to fetch web content.
  • Ingestion points: infsh app run tavily/search-assistant in SKILL.md.
  • Boundary markers: Absent in the example commands.
  • Capability inventory: Access to the infsh CLI tool and network operations via apps.
  • Sanitization: None detected for the data returned from searches before it is likely used for post generation.
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 19, 2026, 03:41 AM