The Agent Skills Directory

[PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection. It ingests data from untrusted design files (.kicad_sch, .kicad_pcb, PDFs) and processes them into structured JSON. If an attacker embeds malicious instructions into component values or labels, the AI agent could be influenced to perform unintended actions when using the skill's capabilities, such as creating GitHub issues or making API requests.
[COMMAND_EXECUTION]: The export_issues.py script uses subprocess.run to call the GitHub CLI (gh) for issue creation and authentication status. what_if.py also uses subprocess.run to chain internal suite tools. These calls use list-based arguments, which is a safe practice that prevents shell injection.
[EXTERNAL_DOWNLOADS]: The lifecycle_audit.py script makes outbound network connections to well-known electronics distributor APIs (DigiKey, Mouser, LCSC, and element14) using urllib.request.urlopen. These connections are used to fetch component lifecycle and technical data based on Manufacturer Part Numbers (MPNs) extracted from the design.

kicad