Scan the portals, score each listing against a structured rubric, tailor the CV, track what you sent. The whole loop stays inside the CLI you already have open.
A letter grade that collapses to a number on a fixed 1.0 to 5.0 range is a thing you can sort. An impression is not.
The README names Claude Code, Codex, OpenCode and Antigravity. The tool assumes you already picked one.
Your CV and your rejections stay on your machine. For this particular corpus that matters more than usual.
Concept mock, not affiliated with the project. The A-F scale and the 1.0-5.0 range are quoted from the repository's own description; the highlighted grade and the four numbered steps are marked illustrative because no sample scoring output is published on the trending page.