The Agent Skills Directory

[COMMAND_EXECUTION]: The script fetch_datasheet_lcsc.py uses subprocess.run to execute the pdftotext utility for verifying the content of downloaded PDFs. Additionally, sync_datasheets_lcsc.py invokes a local Python script analyze_schematic.py to extract component data from KiCad files. These executions use argument lists rather than shell strings, which significantly reduces the risk of command injection.
[EXTERNAL_DOWNLOADS]: The skill fetches electronic component data from jlcsearch.tscircuit.com and downloads datasheet PDFs from wmsc.lcsc.com (LCSC's official CDN) and microchip.com. These sources are well-known and legitimate services within the electronics engineering domain.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it processes untrusted external data from KiCad schematic files and MPN (Manufacturer Part Number) lists.
Ingestion points: The sync_datasheets_lcsc.py script accepts paths to .kicad_sch files and text-based MPN lists provided by the user.
Boundary markers: The skill does not explicitly use boundary markers for LLM instructions within the processed data, as the data is primarily handled by the Python scripts.
Capability inventory: The skill has the capability to write files (PDFs and manifest.json), perform network GET requests, and execute specific subprocesses (pdftotext and the schematic analyzer).
Sanitization: The skill implements robust sanitization through sanitize_filename to prevent path traversal and is_real_mpn (a regex-based filter) to ensure only valid-looking part numbers are processed for searches and file naming.

lcsc