Laniameda Brand Design

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill includes an instruction to source all textual content from an external file (lib/content.ts), creating a surface for indirect prompt injection.\n
  • Ingestion points: lib/content.ts (referenced in SKILL.md rules).\n
  • Boundary markers: None specified for the external content interpolation.\n
  • Capability inventory: Access to Pencil MCP tools including batch_design, get_screenshot, and open_document.\n
  • Sanitization: No sanitization or validation steps are defined for the ingested content.\n- [SAFE]: Command-line references (e.g., npx @laniameda/agent-hub) are used exclusively as mockup text for terminal UI components and do not constitute executable instructions or unauthorized downloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 02:51 PM