One command. 8 job boards searched, companies researched, resume tailored, cover letter written, interviews prepped, offers negotiated — and applications filed. Works for any role, any industry, worldwide.
JobPulse works for any role, any seniority, any location. Tell it what you're looking for — it does the rest.
Engineers, architects, PMs, data scientists. Multi-board scraping with stack-matching scorer and ATS keyword alignment.
Sales, partnerships, growth, strategy. Domain scorers tuned for BD signals — startup, venture, GTM, channel partners.
Space tech, deep tech, blockchain/Web3, climate, biotech, fintech. Custom signal libraries score roles in thin markets others miss.
Pivoting industries? The resume builder extracts transferable skills, the customizer bridges vocabulary gaps, the research brief tells you what matters.
Requires Claude Code, Python ≥ 3.10, Node.js ≥ 18. No accounts, no API keys.
Inside Claude Code, run these two commands. Dependencies install automatically in the background — Python venv, JobSpy, pandas.
Or clone and run locally:
claude --plugin-dir /path/to/jobpulse-ai-agent
Fully quit and reopen Claude Code. The plugin registers on startup and runs a one-time dependency check — about 30 seconds on first launch.
You'll see [jobfinder] Dependencies ready. in the startup log when it's done.
Type the slash command or describe what you want in plain language. The agent guides you through the rest — one question at a time, no form to fill.
Or naturally: "Find remote product manager jobs in Europe, fintech preferred"
job-search/ inside it)
Manual job searching is fragmented, repetitive, and slow. JobPulse handles the entire pipeline end-to-end.
| Feature | JobPulse | Manual search | LinkedIn Premium | Generic AI (ChatGPT) |
|---|---|---|---|---|
| Multi-board search (8 boards at once) | ✓ | One board at a time | LinkedIn only | — |
| AI-tailored resume per job | ✓ Auto-generated | Manual, hours per job | — | Manual prompting |
| Targeted cover letter | ✓ Auto-generated | Manual | — | Manual prompting |
| Company research brief | ✓ Structured | Manual googling | Partial | Manual prompting |
| Interview prep (STAR outlines) | ✓ Role-specific | — | — | Manual prompting |
| Salary negotiation strategy | ✓ Market-backed | — | Salary data only | Manual prompting |
| Browser-assisted application | ✓ With confirmation | — | — | — |
| Data privacy (100% local) | ✓ No cloud sync | ✓ | Stored on LinkedIn | Sent to OpenAI |
| Cost | $0 | $0 | ~$40/mo | ~$20/mo |
No commands to memorize. Talk to it like a recruiter. The agent handles routing, search, and delivery.
"Find remote senior PM roles in Europe, fintech or climate preferred"
Runs 7+ query variations, scores by relevance, tiered shortlist in ~60s.
"Find blockchain business development roles in Europe, startup experience required"
Domain-specific scorer — tuned for Web3 BD signals across 8 boards.
"Review my resume and tell me what's hurting my chances"
Section-by-section score, pitfalls checklist, improved version saved locally.
"What's the culture like at Revolut? Any red flags?"
Browses Glassdoor, LinkedIn, Crunchbase, GitHub. Structured brief with red/green flags.
"They offered me €85k. Help me counter-offer with market data"
Researches P25/median/P75 rates, drafts counter email, provides BATNA talking points.
"Optimize my LinkedIn headline and About section for recruiter search"
Audits each section, rewrites headline + summary, suggests high-value keywords.
"Write a cover letter for this Stripe PM role — max 250 words"
3 paragraphs, company-specific opener from research, real metrics from your resume.
"Apply to Monzo for the product manager position"
Opens form, fills fields with your confirmation, uploads PDF, submits only when you say so.
Every output saved locally inside job-search/[company]/
Culture, funding, tech stack, red/green flags, Glassdoor patterns, and Resume Customization Notes for the agent pipeline.
research.mdYour base resume re-keyed to the JD vocabulary and reordered by relevance. Factual — no invented experience.
[name]-resume-[company].pdf3 paragraphs, 200–280 words. Company-specific opener, real metrics from your resume, zero clichés.
cover-letter.md5–6 behavioral questions with STAR outlines drawn from your actual resume. Technical Qs and smart questions to ask them.
interview-prep.mdMarket rate data (P25/median/P75), offer evaluation, counter email draft, and BATNA talking points.
salary-negotiation.mdAudit score per section, optimized headline formula, rewritten About section, skills gap analysis, completeness checklist.
linkedin-optimized.mdDirect apply URL, source board, date found, and application status updated automatically when you apply.
apply-link.mdOpens the form, fills section by section, uploads your customized PDF. Submits only on your explicit "submit" — never auto-submits.
/apply skillEverything you need to know before getting started.
job-search/ folder on your own machine.
Nothing is uploaded to any server. No telemetry, no cloud sync, no accounts. The only data that leaves your machine
are the search queries sent to job boards (same as opening a browser tab).
country_indeed="sweden") are set automatically based on your stated location preference.
Note: Glassdoor and ZipRecruiter are US/CA focused; for European searches the agent skips them automatically.
Free, open source, 8 boards, 8 agents — everything on your machine.
Apache 2.0 · No telemetry · No cloud sync · Your data stays yours