skills/nomadamas/k-skill/saju-fortune/Gen Agent Trust Hub

saju-fortune

Warn

Audited by Gen Agent Trust Hub on Aug 1, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the saju-fortune package from the NPM registry at runtime if it is missing.
  • [COMMAND_EXECUTION]: The skill uses shell commands to install global packages and execute scripts via the node CLI.
  • [REMOTE_CODE_EXECUTION]: The skill uses a heredoc pattern to pipe dynamically generated JavaScript code into a Node.js process. This code includes user-provided variables like names and locations, which can lead to code injection if malicious inputs are provided.
  • [DATA_EXFILTRATION]: The skill requests sensitive personally identifiable information (PII) such as the user's name, birth date, exact birth time, and birth city.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection or command injection because it interpolates user inputs directly into script templates without validation or escaping. 1. Ingestion points: User interview inputs (name, birthDate, birthTime, gender, birthCity) in SKILL.md. 2. Boundary markers: Absent in code templates. 3. Capability inventory: Global package installation, shell execution of Node.js scripts. 4. Sanitization: No sanitization or escaping of user strings is present before interpolation.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 1, 2026, 05:17 AM
Security Audit — agent-trust-hub — saju-fortune