pipeline

To automate job posting URL processing, this Skill fetches pending links, extracts job descriptions using Playwright or WebFetch, and applies an A–G rubric to generate scored evaluation reports and tracker entries automatically.

Login-gated URLs are handled safely by skipping them during web scraping. The pipeline marks inaccessible job posting links with an actionable note in the tracker output rather than failing the extraction.

You can use Playwright, WebFetch, and WebSearch together for job description extraction. The pipeline attempts Playwright first, falls back to WebFetch, then WebSearch to ensure robust content retrieval before scoring.

To generate PDFs and tracker entries, the pipeline processes pending URLs, applies an A–G score threshold, creates the next sequential report number, outputs a PDF, and writes the outcome into the designated applications tracker path.

Pipeline inbox processing requires checking sync states via the cv-sync-check script before extracting job descriptions. The workflow reads pipeline.md pending and processed states to maintain structured status markers and sequential reporting numbers.

Job description extraction fails when URLs are login-gated or inaccessible. The pipeline handles these safely by skipping the posting and logging an actionable note, ensuring the scoring workflow continues processing remaining pending entries without interruption.