youmind-webpage-generator

Warn

Audited by Gen Agent Trust Hub on Mar 17, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The workflow in SKILL.md (Step 3) performs direct string interpolation of user-provided descriptions into a youmind call shell command. This creates a command injection vulnerability where shell metacharacters or single quotes in the description could be exploited to execute arbitrary commands on the host system.\n- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the @youmind-ai/cli package. While this is a vendor-owned resource, global installation of external binaries increases the attack surface of the environment.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) due to the ingestion of untrusted data into a high-capability workflow.\n
  • Ingestion points: User description input processed in Step 3 of SKILL.md.\n
  • Boundary markers: None; the description is interpolated directly into a JSON payload within a shell command.\n
  • Capability inventory: Bash command execution, vendor API interaction via CLI, and Node.js execution.\n
  • Sanitization: No validation or escaping is applied to the description before interpolation.\n- [COMMAND_EXECUTION]: Step 5 of the workflow uses node -e to execute a script for parsing JSON results. Although the script logic appears benign, the use of dynamic execution to handle external data is a security-sensitive practice.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 17, 2026, 01:58 PM