Skills
skills/1nfsh/skills/linkedin-content/Gen Agent Trust Hub

linkedin-content

Fail

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • REMOTE_CODE_EXECUTION (CRITICAL): The skill uses a dangerous pattern to download and execute code from an untrusted source: curl -fsSL https://cli.inference.sh | sh. This allows for arbitrary code execution on the user's machine without verification of the script's contents.
  • EXTERNAL_DOWNLOADS (HIGH): The skill downloads and installs an unverified CLI tool and additional skills from the inference-sh repository, which is not a trusted provider.
  • COMMAND_EXECUTION (MEDIUM): The skill enables and utilizes the infsh command-line tool via the Bash tool, allowing execution of complex operations based on instructions from an unverified remote source.
  • PROMPT_INJECTION (LOW): Indirect Prompt Injection Surface identified. Evidence Chain: 1. Ingestion points: Untrusted data enters the agent context via infsh app run tavily/search-assistant in SKILL.md. 2. Boundary markers: Absent (no delimiters or warnings found). 3. Capability inventory: The skill possesses Bash(infsh *) capabilities in SKILL.md. 4. Sanitization: Absent (no evidence of escaping or validation).
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 18, 2026, 01:23 AM
Security Audit — agent-trust-hub — linkedin-content