ediq 2.0.0

Official Python SDK for Ediq AI Detection API - Education & HR modes

Ediq - AI Detection SDK

Official Python client for Ediq AI Detection API by Wyzcon.

Detect AI-generated content in educational documents (essays, assignments) and HR documents (resumes, cover letters, LinkedIn profiles) with 96% accuracy.

Installation

pip install ediq

Quick Start

from ediq import Ediq

client = Ediq("wyz_xxxxxxxxxxxxxxxxxxxxxxxxxxxxx")

# Education: Detect AI in student essays
result = client.detect_edu("Student essay text...")
print(f"AI Probability: {result.probability}%")

# HR: Detect AI in resumes
result = client.detect_hr("Resume text...", context="resume")
print(f"AI Probability: {result.probability}%")

Get Your API Key

1. Sign up at wyzcon.com
2. Get your free API key
3. Start detecting AI content

Two Detection Modes

🎓 Education Mode

For essays, assignments, and student work. Supports student baselines for personalized detection.

💼 HR Mode

For resumes, cover letters, and LinkedIn profiles. Optimized for professional document patterns.

Features

• ✅ 96% accuracy on 10,000+ tested documents
• ✅ 9-layer detection system
• ✅ Student baselines - Protect good writers (Education)
• ✅ HR context awareness - Resume, cover letter, LinkedIn (HR)
• ✅ Photo scanning - OCR for handwritten work
• ✅ Comprehensive reports - Detailed analysis
• ✅ Type hints - Full autocomplete support

Education Mode Examples

Basic Education Detection

from ediq import Ediq

client = Ediq("your_api_key")

# Analyze student essay
result = client.detect_edu("The importance of renewable energy...")
print(f"AI: {result.probability}%")
print(f"Assessment: {result.assessment}")

With Student Baseline

Compare against a student's verified previous work to reduce false positives:

result = client.detect_edu(
    text="Current submission...",
    student_id="john_doe",
    baseline="Previous authentic work..."
)

print(f"AI: {result.probability}%")
print(f"Baseline Similarity: {result.baseline_similarity:.1%}")

Detect from File (Education)

result = client.detect_edu_file("essay.pdf")
print(f"AI: {result.probability}%")

Detect from Image (Handwritten)

result = client.detect_edu_image(
    "handwritten_essay.jpg",
    handwritten=True
)
print(f"AI: {result.probability}%")

Formal Writing Mode

For academic papers and formal essays (reduces false positives):

result = client.detect_edu(
    text="Formal academic paper...",
    formal_mode=True
)

HR Mode Examples

Detect AI in Resume

from ediq import Ediq, HRContextType

client = Ediq("your_api_key")

# Analyze resume text
result = client.detect_hr(
    text="Experienced software engineer with 5+ years...",
    context="resume"  # or HRContextType.RESUME
)

print(f"AI: {result.probability}%")
print(f"Assessment: {result.assessment}")

Detect AI in Cover Letter

result = client.detect_hr(
    text="I am writing to express my interest...",
    context="cover_letter"
)
print(f"AI: {result.probability}%")

Detect AI in LinkedIn Profile

result = client.detect_hr(
    text="Passionate technology leader driving innovation...",
    context="linkedin_profile"
)
print(f"AI: {result.probability}%")

Detect from HR File

# Analyze resume PDF
result = client.detect_hr_file(
    "resume.pdf",
    context="resume"
)
print(f"AI: {result.probability}%")

# Analyze cover letter DOCX
result = client.detect_hr_file(
    "cover_letter.docx",
    context="cover_letter"
)

With Comprehensive Report

result = client.detect_hr(
    text="Resume text...",
    context="resume",
    include_report=True
)

print(result.report)  # Detailed writing analysis

HR Context Types

Context	Use For
resume	CV, resume documents
cover_letter	Job application cover letters
linkedin_profile	LinkedIn about/summary sections
other	Other professional documents

Check Usage

usage = client.usage()
print(f"Used: {usage.used}/{usage.limit}")
print(f"Remaining: {usage.remaining}")
print(f"Tier: {usage.tier}")

DetectionResult Properties

Property	Type	Description
probability	float	AI probability (0-100)
assessment	str	Category: likely_human, borderline, suspicious, likely_ai, highly_likely_ai
confidence	float	Confidence score (0-100)
word_count	int	Words analyzed
mode	str	Detection mode: "education" or "hr"
scan_id	int	Saved scan ID
baseline_similarity	float	Baseline match (Education only)
hr_context	str	Document context (HR only)
report	dict	Full writing analysis (if requested)
layers	dict	Layer-by-layer breakdown

Error Handling

from ediq import Ediq, EdiqError, RateLimitError, AuthenticationError

client = Ediq("your_api_key")

try:
    result = client.detect_edu("Text...")
except AuthenticationError:
    print("Invalid API key")
except RateLimitError:
    print("Rate limit exceeded - wait and retry")
except EdiqError as e:
    print(f"API error: {e}")

Migration from v1.x

The old methods still work but are deprecated:

# Old (still works, defaults to education mode)
result = client.detect("Text...")
result = client.detect_file("essay.pdf")
result = client.detect_image("photo.jpg")

# New (explicit mode selection)
result = client.detect_edu("Text...")
result = client.detect_hr("Text...", context="resume")

Documentation

Full documentation at docs.wyzcon.com

License

MIT License