skills/firzus/agent-skills/artefact/Gen Agent Trust Hub

artefact

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill incorporates external libraries from well-known services: Tailwind CSS from cdn.tailwindcss.com and Mermaid.js from cdn.jsdelivr.net for rendering visualizations.\n- [COMMAND_EXECUTION]: To present the visualization to the user, the skill executes platform-specific system commands (xdg-open, open, start, or explorer.exe) to launch the generated HTML file in the default web browser.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it transforms 'concept' data—which may be sourced from untrusted external inputs—into an HTML document. The Mermaid.js configuration is set to securityLevel: "loose", which permits the rendering of HTML tags and potentially malicious scripts within diagrams, posing a Cross-Site Scripting (XSS) risk in the browser context.\n
  • Ingestion points: Processes arbitrary 'concept' data provided at runtime to populate the HTML template.\n
  • Boundary markers: No delimiters or warnings are used to prevent the AI from obeying instructions embedded in the 'concept' data.\n
  • Capability inventory: Writes to the local filesystem (temp directory) and executes system commands to open files.\n
  • Sanitization: The skill does not specify any sanitization or escaping mechanisms for the data before it is written into the HTML file.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 10:16 PM
Security Audit — agent-trust-hub — artefact