The resume-intel command-line interface — parse resume PDFs directly from the terminal.

CLI

resume-intel ships a command-line interface that lets you parse resume PDFs directly from the terminal without writing any code.

Installation

The CLI is included when you install the package:

npm install -g @edwinfom/resume-intel ai @ai-sdk/deepseek
# or use npx without installing globally
npx @edwinfom/resume-intel parse resume.pdf

Usage

resume-intel parse <file.pdf> [options]

Options

Option Description Default
--model <provider:model> AI model to use deepseek:deepseek-chat
--sections <s1,s2,...> Sections to extract 8 default sections
--layout <spatial|linear> Layout strategy spatial
--output <json|pretty> Output format pretty
--out <file.json> Write output to file stdout
--no-decompose Use single-shot extraction
--key <api-key> API key (or use env variable)
--help Show help

Models

# DeepSeek — best cost/performance (default)
# Requires: DEEPSEEK_API_KEY env var or --key
resume-intel parse resume.pdf --model deepseek:deepseek-chat
 
# OpenAI
# Requires: OPENAI_API_KEY env var or --key
resume-intel parse resume.pdf --model openai:gpt-4o-mini
 
# Anthropic
# Requires: ANTHROPIC_API_KEY env var or --key
resume-intel parse resume.pdf --model anthropic:claude-3-5-haiku-20241022
 
# Ollama — local, no API key needed
# Requires: ollama serve running locally
resume-intel parse resume.pdf --model ollama:llama3.1

Examples

# Parse with DeepSeek (API key from environment)
DEEPSEEK_API_KEY=sk-... resume-intel parse resume.pdf
 
# Parse and write JSON to file
resume-intel parse resume.pdf --out result.json
 
# Extract only basics and work experience
resume-intel parse resume.pdf --sections basics,work
 
# Extract all 12 sections
resume-intel parse resume.pdf --sections basics,work,education,skills,languages,projects,awards,certificates,publications,volunteer,interests,references
 
# Raw JSON output for piping
resume-intel parse resume.pdf --output json | jq .data.basics.name
 
# Use OpenAI with explicit key
resume-intel parse resume.pdf --model openai:gpt-4o-mini --key sk-...
 
# Single-shot extraction (faster for simple CVs)
resume-intel parse resume.pdf --no-decompose

Pretty output

By default, the CLI prints a formatted summary to the terminal:

resume-intel v0.2.0 — Extraction Result
------------------------------------------------------------

  BASICS
    Name:     Jane Doe
    Title:    Senior Software Engineer
    Email:    jane@example.com
    Location: San Francisco, California, US
    Profiles: LinkedIn (janedoe) / GitHub (janedoe)

  WORK EXPERIENCE  (2)
    Staff Engineer @ Acme Corp  2021-03 -> Present
      - Led migration of monolith to microservices
      - Managed a team of 8 engineers
    Senior Engineer @ Startup Inc  2018-06 -> 2021-02

  EDUCATION  (1)
    MIT  BSc in Computer Science  2010 -> 2014

  SKILLS
    Backend: Node.js, TypeScript, Go, PostgreSQL
    Infrastructure: Kubernetes, Terraform, AWS

  LANGUAGES
    English (Native)  /  French (Conversational)

  PROJECTS  (1)
    open-source-tool
      A CLI tool for...
      https://github.com/janedoe/open-source-tool

------------------------------------------------------------
  METADATA
    Duration:  2340ms
    Pages:     2
    OCR:       no (text-native)
    Sections:  basics, work, education, skills, languages, projects, certificates, volunteer
    Tokens:    1850 total  (1200 prompt + 650 completion)
------------------------------------------------------------

Environment variables

Variable Provider
DEEPSEEK_API_KEY DeepSeek
OPENAI_API_KEY OpenAI
ANTHROPIC_API_KEY Anthropic
GOOGLE_API_KEY Google Gemini