CLInkedIn

CLInkedIn is a self-hosted LinkedIn jobs scanner with two frontends backed by the same API:

• a web dashboard,
• a colorful terminal dashboard (TUI/CLI).

Includes hourly auto-scanning, and a unified feed from parallel keyword/location targets.

Features

Feed and UX

• Unified jobs feed sorted by freshest posting age.
• Default filter shows only jobs posted in the last 3 days.
• Quick toggles for all jobs and starred-only view.
• Fast full-text search by title, company, location.
• One-click "Open on LinkedIn" action per row.

Job workflow

• Star/unstar jobs and highlight starred rows.
• Persisted job exclusions and company exclusions.
• Exclusions are applied to future scans and feed rendering.

Scanning and automation

• Hourly scheduler in the web service.
• Manual scan trigger from web and CLI.
• Parallel scanners for multiple keyword/location targets.
• Unified deduped feed across all enabled scan targets.

Intelligence and enrichment

• Optional Groq-powered keyword expansion/refinement (free API key from their site)
• Web logo matching/search pipeline with scoring by source, filename, filetype, and domain/name similarity.

Interfaces

• Web UI: browser dashboard for feed + config + exclusions.
• CLI/TUI: non-flicker terminal UI with keyboard navigation, auto-refresh, and inline config editing.

Credits

• Scraper baseline/reference: luminati-io/LinkedIn-Scraper

Quick Start (Docker Compose)

docker compose up -d --build

Open web UI:

• http://<your-server-ip>:8765

Run CLI Alongside Web

Local

./.venv/bin/python job_scanner_cli.py --api-base-url http://127.0.0.1:8765

Docker Compose profile

docker compose --profile cli run --rm linkedin-job-cli

or just use

./run.sh cli

CLI keys:

• esc escape search
• r run scan
• f cycle feed filter (recent / all / starred)
• s star/unstar selected row
• x exclude selected job
• c exclude selected company
• g open JSON config editor
• / focus search
• u refresh now
• q quit

Development

Prereqs

• Python 3.12+ (project also works in current local .venv)
• Docker + Docker Compose (for containerized run)

Local setup

./.venv/bin/pip install -r requirements-dev.txt

Run API + web locally

./.venv/bin/uvicorn app.main:app --host 0.0.0.0 --port 8765

Run tests

./.venv/bin/python -m pytest -q

Configuration

Primary runtime files (persisted in ./data):

• data/linkedin_jobs.json
• data/app_state.json
• data/scan_config.json

Main env vars:

• APP_PORT (default 8765)
• SCAN_INTERVAL_MINUTES (default 60)
• SCAN_ON_STARTUP (default true)
• SCAN_KEYWORDS, SCAN_LOCATION, SCAN_MAX_JOBS
• STATE_JSON_PATH, SCAN_CONFIG_PATH, JOBS_JSON_PATH
• GROQ_API_KEY, GROQ_MODEL
• LOGO_EXTERNAL_SEARCH_ENABLED (default true)

Public Repo Hygiene

• Secrets are excluded by .gitignore (.env, .env.*).
• Use .env.example as your local template.
• Runtime JSON state and job data are excluded from git.