kicad-schematic

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill guides the agent to perform internet searches and extract data from external PDFs and datasheets. This creates a surface for indirect prompt injection where adversarial content within a downloaded datasheet could try to influence the agent's actions.\n
  • Ingestion points: Ingesting external PDFs, datasheets, and reference designs from arbitrary internet sources.\n
  • Boundary markers: Absent. No clear instructions or boundaries are provided to isolate or ignore potential text commands within the downloaded datasheets.\n
  • Capability inventory: Modifying local KiCad project files, creating or editing schematic symbols, and executing web searches.\n
  • Sanitization: Absent. There is no mention of filtering or sanitizing the content extracted from external documents.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 10:08 PM
Security Audit — agent-trust-hub — kicad-schematic