kicad-footprint

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends cloning the official KiCad Library Tools repository from its GitLab location (https://gitlab.com/kicad/libraries/kicad-library-tools) to provide the necessary generator scripts.
  • [INDIRECT_PROMPT_INJECTION]: The skill establishes a surface for indirect prompt injection by processing instructions and context from project files and user-supplied content.
  • Ingestion points: The agent is instructed to read project-specific files such as AGENTS.md or CLAUDE.md and process user-provided PDF/image descriptions of land patterns.
  • Boundary markers: No specific delimiters or instruction-isolation markers are provided to protect the agent from embedded commands in these files.
  • Capability inventory: The agent is permitted to read local library definitions, generate new YAML files, and interact with the footprint generation infrastructure.
  • Sanitization: No explicit sanitization or validation logic is defined for the ingested context, although human review is suggested for non-standard footprint shapes.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 07:06 PM
Security Audit — agent-trust-hub — kicad-footprint