ai-marketing-videos

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 contains the command curl -fsSL https://cli.inference.sh | sh, which downloads and executes a script from a non-trusted domain. This pattern is a high-risk security violation as the contents of the script are unverified and can be modified at the source without notice.\n- [EXTERNAL_DOWNLOADS] (MEDIUM): The skill references and installs multiple external packages via npx skills add inference-sh/skills. These sources are not included in the trusted organizations list, making them unverifiable dependencies.\n- [COMMAND_EXECUTION] (MEDIUM): The skill relies on the Bash tool to interact with the infsh CLI. This provides the agent with the ability to execute persistent commands, write files to the system (e.g., > hook.json), and perform network operations.\n- [PROMPT_INJECTION] (LOW): The skill is vulnerable to indirect prompt injection (Category 8). 1. Ingestion points: User-provided prompts are interpolated into shell commands. 2. Boundary markers: None are present in the provided templates to prevent prompt escape. 3. Capability inventory: Includes file-system writing and external tool execution via Bash. 4. Sanitization: No visible input sanitization or validation of the prompt strings is implemented.
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, 01:50 AM
Security Audit — agent-trust-hub — ai-marketing-videos