Use the same parser in a terminal or review workflow.

Run from the command line

The CLI uses the same scanner core as the browser. It accepts PDF, DOCX, TXT, and Markdown files and performs no network request.

npx ats-readback resume.pdf
npx ats-readback resume.docx --job job-description.txt
npx ats-readback resumes/ --recursive --format json --output reports.json

Before the npm package is published, run the repository directly:

npx github:hassanalshama/ats-readback resume.pdf

Options

ats-readback [options] <file-or-directory> [...]

-j, --job <path>          compare with a job-description text file
-f, --format <name>       text, json, markdown, or html
-o, --output <path>       write output to a file
-r, --recursive           scan supported files inside directories
    --fail-on <severity>  high, medium, low, info, or never
    --max-size <value>    per-file limit, such as 50MB
    --stdin-name <name>   filename used for standard input
    --no-color            disable terminal colors
-q, --quiet               print only the verdict line

Use in a review workflow

The exit code is designed for scripts:

ats-readback deliverables/   --recursive   --job role.txt   --format markdown   --output ats-readback.md   --fail-on medium

Report formats

Text is designed for terminal review. JSON preserves every finding, extraction metric, contact value, section, term match, and coverage limitation. Markdown works well in pull requests. HTML is a portable, self-contained report.

Reports contain recovered resume text. Do not publish them by default.

Local development

git clone https://github.com/hassanalshama/ats-readback.git
cd ats-readback
npm run verify
npm run dev

The project has zero runtime dependencies. npm run verify regenerates deterministic fixtures, checks source and links, runs the tests, builds the static site, and validates the npm package.