kicad-placement
Pass
Audited by Gen Agent Trust Hub on Nov 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external information from datasheets and floor plans to determine PCB layout. A malicious datasheet could contain instructions designed to influence the agent's behavior or manipulate the generated placement script.
- Ingestion points: External datasheet layout sections and floor plan requirements.
- Boundary markers: None specified in the instructions to isolate external data.
- Capability inventory: Writing and executing Python scripts (board.py), using KiCad editing tools (kicad-board-edit), and saving board changes (save()).
- Sanitization: No sanitization or verification procedures are outlined for processing external datasheet content.
- [DYNAMIC_EXECUTION]: The skill instructs the agent to create, maintain, and execute a local Python script (board.py) to automate placement tasks.
- Evidence: The instructions direct the agent to manage a placement script containing generated function calls (e.g., place(ref, x, y, rot)) and to rerun this script whenever changes are made.
Audit Metadata