career-ops MCP — Anwendungsfälle, Installieren & Live-Demo

Warum nutzen

Hauptfunktionen

14 skill modes covering every stage
Resume-to-JD matching with keyword coverage scoring
Cover letter drafting with your voice profile
PDF generation (LaTeX-quality from Markdown)
Go dashboard to track application status

Live-Demo

In der Praxis

bereit

Installieren

Wählen Sie Ihren Client

~/Library/Application Support/Claude/claude_desktop_config.json · Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "career-ops-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/santifer/career-ops",
        "~/.claude/skills/career-ops"
      ],
      "_inferred": true
    }
  }
}

Öffne Claude Desktop → Settings → Developer → Edit Config. Nach dem Speichern neu starten.

~/.cursor/mcp.json · .cursor/mcp.json

{
  "mcpServers": {
    "career-ops-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/santifer/career-ops",
        "~/.claude/skills/career-ops"
      ],
      "_inferred": true
    }
  }
}

Cursor nutzt das gleiche mcpServers-Schema wie Claude Desktop. Projektkonfiguration schlägt die globale.

VS Code → Cline → MCP Servers → Edit

{
  "mcpServers": {
    "career-ops-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/santifer/career-ops",
        "~/.claude/skills/career-ops"
      ],
      "_inferred": true
    }
  }
}

Klicken Sie auf das MCP-Servers-Symbol in der Cline-Seitenleiste, dann "Edit Configuration".

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "career-ops-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/santifer/career-ops",
        "~/.claude/skills/career-ops"
      ],
      "_inferred": true
    }
  }
}

Gleiche Struktur wie Claude Desktop. Windsurf neu starten zum Übernehmen.

~/.continue/config.json

{
  "mcpServers": [
    {
      "name": "career-ops-skill",
      "command": "git",
      "args": [
        "clone",
        "https://github.com/santifer/career-ops",
        "~/.claude/skills/career-ops"
      ]
    }
  ]
}

Continue nutzt ein Array von Serverobjekten statt einer Map.

~/.config/zed/settings.json

{
  "context_servers": {
    "career-ops-skill": {
      "command": {
        "path": "git",
        "args": [
          "clone",
          "https://github.com/santifer/career-ops",
          "~/.claude/skills/career-ops"
        ]
      }
    }
  }
}

In context_servers hinzufügen. Zed lädt beim Speichern neu.

claude mcp add career-ops-skill -- git clone https://github.com/santifer/career-ops ~/.claude/skills/career-ops

Einzeiler. Prüfen mit claude mcp list. Entfernen mit claude mcp remove.

Anwendungsfälle

Praxisnahe Nutzung: career-ops

Apply to 30 roles with tuned resumes and cover letters in an afternoon

👤 Job seekers, especially mid-career switching roles ⏱ ~180 min intermediate

Wann einsetzen: You've saved 30 JDs and each wants a slightly different resume angle.

Voraussetzungen

Skill installed — git clone https://github.com/santifer/career-ops ~/.claude/skills/career-ops
Base resume in Markdown — Convert current resume to /career/resume-base.md

Ablauf

Ingest JDs

Use career-ops. Load JDs from /career/jds/. For each, score my resume match + identify the 3 highest-leverage tweaks.✓ Kopiert

→ Scorecard per JD with concrete tweak suggestions
Tune + cover letter

For each JD: produce a tuned resume + cover letter. Save to /career/applications/<company>/✓ Kopiert

→ Per-company folder with both docs
Render PDFs

Render every .md to PDF. Verify page count <= 2 for resumes.✓ Kopiert

→ PDFs generated; page counts reported

Ergebnis: A stack of tuned application packets, ready to submit in batch.

Fallstricke

Over-tuning — resumes look inconsistent across applications — The skill preserves your narrative spine; review a few manually

Kombinieren mit: filesystem

Run the Go dashboard to track applications

👤 Job seekers needing kanban-style visibility ⏱ ~15 min beginner

Wann einsetzen: You've applied to 30+ places and lost track of who's at what stage.

Voraussetzungen

Go installed — brew install go or grab from go.dev

Ablauf

Start dashboard

Use career-ops to start the dashboard on localhost:7070.✓ Kopiert

→ Dashboard visible with current apps
Update status

Move "Acme Co" to "interview-scheduled" for April 30 10am.✓ Kopiert

→ Card updated on board
Weekly review

What's stale (no action in 14 days)? What's my response rate?✓ Kopiert

→ Stats + stale-item list

Ergebnis: A working pipeline view, not a mental juggle.

Fallstricke

Dashboard DB isn't backed up — Add the DB path to your dotfiles sync or back it up manually

Kombinationen

Mit anderen MCPs für 10-fache Wirkung

career-ops-skill + filesystem

Organize per-company folders

Save all applications under /career/applications/<slug>/ with a standard structure.✓ Kopiert

career-ops-skill + humanizer-skill

Make cover letters not sound AI

Run humanizer on the cover letter before rendering PDF.✓ Kopiert

Werkzeuge

Was dieses MCP bereitstellt

Werkzeug	Eingaben	Wann aufrufen
score_jd_match	resume, jd	Per-JD triage
tune_resume	base, jd, tweaks	Once you commit to applying
draft_cover_letter	resume, jd, voice_samples	Always after resume
render_pdf	md_path	Final output stage
dashboard_start	port?	Tracking mode

Kosten & Limits

Was der Betrieb kostet

API-Kontingent: N/A
Tokens pro Aufruf: High in batch mode — expect significant token use per batch
Kosten in €: Free (MIT)
Tipp: Batch 5 at a time, not 30 — you'll catch voice drift and recalibrate

Sicherheit

Rechte, Secrets, Reichweite

Minimale Scopes: filesystem-read filesystem-write

Credential-Speicherung: Dashboard DB local

Datenabfluss: None

Your resume contains PII — don't share outputs publicly

Fehlerbehebung

Häufige Fehler und Lösungen

PDF rendering fails

LaTeX or Typst must be installed. Skill tells you which path it uses — install accordingly.

Dashboard port in use

Pass --port 7171

Prüfen: `lsof -i :7070`

Cover letters all sound identical

Per-company personalization is on by default; increase the per-JD research step depth

Alternativen

career-ops vs. andere

Alternative	Wann stattdessen	Kompromiss
Huntr / Teal	You want a polished SaaS job tracker	Not LLM-integrated; manual data entry
A plain spreadsheet + Word template	You're only applying to a few roles	No automation on the tuning side

Mehr

Ressourcen

📖 Offizielle README auf GitHub lesen

🐙 Offene Issues ansehen

🔍 Alle 400+ MCP-Server und Skills durchsuchen