The Agent Skills Directory

[DYNAMIC_EXECUTION]: The check.py script dynamically modifies the Python module search path using sys.path.insert(0, os.path.join(HERE, '..', 'kicad-schematic')) to import a lint function from a sibling folder. This constitutes dynamic loading from a computed path.
[COMMAND_EXECUTION]: The skill executes external commands via subprocess.run(). It determines the path to the kicad-cli executable by checking the KICAD_CLI environment variable or looking in several hardcoded standard installation paths. This allows execution of a user-provided binary if the environment variable is manipulated.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted KiCad design files (.kicad_sch, .kicad_pcb). \n- Ingestion points: Reads project, schematic, and board files via open().read() and via the kicad-cli tool. \n- Boundary markers: None identified in the prompt templates or processing logic. \n- Capability inventory: File system read access, file copying via shutil.copy, and subprocess execution. \n- Sanitization: There is no sanitization or filtering of text content (such as comments or field values) within the design files before their contents are processed and summarized for the AI agent.

kicad-check