/setup

name: setup
description: One-time setup for Resume Builder — Python dependencies, config.json, cloud or local scoring, and optional LLM features. Use once right after installing, or when the user reports missing dependencies, a missing or incomplete config, wants to switch between cloud and local scoring, or wants to enable AI-augmented scoring with an Anthropic API key.

Resume Builder — One-Time Setup

Run this skill once after installing to get started.

Instructions

You are helping the user set up Resume Builder. Guide them through the steps below.

Step 1: Check Python

Run `python --version` (or `python3 --version` on Mac/Linux).

• If Python 3.10+ is installed, proceed to Step 2.
• If Python is NOT installed, tell the user:

The scoring engine requires Python 3.10 or later.

Download it from: https://www.python.org/downloads/

During installation:
- CHECK "Add Python to PATH" (Windows)
- Mac/Linux: Use your package manager (brew install python3)

After installing Python, run the setup skill again.

Stop here if Python is not installed.

Step 2: Choose Install Type

Ask the user which setup they prefer:

**Option A: Quick Setup (Recommended for most users)**

• Cloud-based scoring — no heavy downloads
• Installs only 2 small packages (~5 seconds)
• Scoring happens on our cloud server (5 free scores, then Pro $12/month for unlimited)
• Optional LLM-augmented scoring with your own Anthropic API key

**Option B: Full Local Setup (For developers / offline use)**

• Installs all scoring engines locally (~500MB, takes 2-3 minutes)
• Scoring runs on your machine — no cloud needed, no limits
• Requires more disk space and RAM

Step 3A: Quick Setup (Cloud)

pip install -r requirements-plugin.txt

If `pip` is not found, try `pip3` or `python -m pip`. If needed, use the `requirements-plugin.txt` file from the installed plugin root.

This installs `fastmcp` (MCP protocol) and `anthropic` (for LLM features). All ATS/HR scoring goes through the cloud API automatically.

Step 3B: Full Local Setup

pip install -r requirements.txt

Use the `requirements.txt` file from the installed plugin root if needed. This takes 1-3 minutes (downloads sentence-transformers model ~80MB).

Then download NLTK data:

python -c "import nltk; nltk.download('wordnet'); nltk.download('punkt_tab')"

Step 4: Set Up Configuration

Check if the user has a `config.json` in their project. If not, create one from `config.example.json`:

1. Read `config.example.json` from the plugin directory 2. Ask the user for their details:

• Full name
• Credentials (e.g., M.D., MBA, CPA — or leave blank)
• Email
• Phone
• LinkedIn URL
• Path to their master resume file (DOCX, PDF, or Markdown with their full work history)

3. Create `config.json` with their answers

Step 5: (Optional) LLM Features Setup

Ask the user if they want to enable advisory AI-powered scoring.

If yes: 1. They need an Anthropic API key from https://console.anthropic.com/ 2. Create a `.env` file with: `ANTHROPIC_API_KEY=sk-ant-...` 3. This enables the advisory `score_with_llm` tool. Legacy direct rewrite tools are not production-authorized; the native Resume Team is the sole production resume rewrite and draft-publication path.

If no, skip this step. The ATS and HR scorers work without an API key.

Step 6: Verify

For Quick Setup, run:

python -c "import fastmcp; print('Plugin ready! Cloud scoring active.')"

For Full Setup, run:

python -c "import ats_scorer; import hr_scorer; print('Full scoring engine ready!')"

If successful, tell the user:

Setup complete! You can now use:

  resume [paste job description]          — Full resume + cover letter package
  tailor-resume [paste JD]                — Resume only
  cover-letter [paste JD]                 — Cover letter only
  writing-coach [resume file]             — Improve resume writing quality

The scoring engine is active and will automatically score your resumes.
Cloud scoring: 5 free scores included. Upgrade at https://getresumehq.com

If it fails, show the error and suggest fixes.