Skills originality-ai

install

source · Clone the upstream repo

git clone https://github.com/TerminalSkills/skills

Claude Code · Install into ~/.claude/skills/

T=$(mktemp -d) && git clone --depth=1 https://github.com/TerminalSkills/skills "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/originality-ai" ~/.claude/skills/terminalskills-skills-originality-ai && rm -rf "$T"

manifest: skills/originality-ai/SKILL.md

safety · automated scan (low risk)

This is a pattern-based risk scan, not a security review. Our crawler flagged:

references API keys

Always read a skill's source content before installing. Patterns alone don't mean the skill is malicious — but they warrant attention.

source content

Originality.ai API Integration

Overview

Originality.ai combines AI content detection with plagiarism checking in a single API. It uses a credit-based pricing model where each scan costs credits proportional to word count.

API base URL:
```
https://api.originality.ai/api/v1
```
Auth:
```
X-OAI-API-KEY
```
header
Docs: https://docs.originality.ai

Instructions

Authentication

Set

ORIGINALITY_API_KEY

in your environment. Pass it in the

X-OAI-API-KEY

header for all requests.

AI Detection — POST

/api/v1/scan/ai

Send

{ "content": "<text>", "aiModelVersion": "1", "storeScan": false }

. Returns:

```
score.ai
```
(0-1) — probability of AI authorship
```
score.original
```
(0-1) — probability of human authorship (equals
```
1 - ai
```
)
```
credits_used
```
— credits consumed by this scan

`score.ai`	Meaning
0.80 - 1.0	Very likely AI-generated
0.50 - 0.79	Mixed / uncertain
0.20 - 0.49	Probably human-written
0.00 - 0.19	Very likely human

Plagiarism Detection — POST

/api/v1/scan/plag

Send

{ "content": "<text>", "storeScan": false }

. Returns:

```
score.percentUnique
```
(0-100) — percentage of unique content
```
score.percentDuplicated
```
(0-100) — percentage matched elsewhere on the web

matches[]

— array of

{ url, matchedWords, percentage }

for each source found

Combined Scan

Run both AI detection and plagiarism checks in parallel for comprehensive content verification. Flag content if AI score >= 0.5 or plagiarism >= 20%.

Credit Management

Check your balance via GET

/api/v1/account/credits/balance

. Credits are consumed per word, and AI + plagiarism scans cost credits separately. Use

storeScan: false

to avoid storing content in the dashboard.

Examples

Example 1: Checking marketing copy for originality

A content team lead verifies a freelancer's submitted blog post about cloud migration:

AI Detection:
POST https://api.originality.ai/api/v1/scan/ai
Headers: { "X-OAI-API-KEY": "oai-key-abc123...", "Content-Type": "application/json" }
Body: { "content": "Cloud migration is a transformative journey that organizations must carefully plan. It is essential to consider the various deployment models available, including public, private, and hybrid cloud solutions. Furthermore, a comprehensive migration strategy should address data security, compliance requirements, and cost optimization.", "aiModelVersion": "1", "storeScan": false }

Response:
{ "success": true, "score": { "ai": 0.92, "original": 0.08 }, "credits_used": 1 }

Plagiarism Check:
POST https://api.originality.ai/api/v1/scan/plag
Body: { "content": "...(same text)...", "storeScan": false }

Response:
{ "success": true, "score": { "percentUnique": 73, "percentDuplicated": 27 }, "matches": [{ "url": "https://example-cloud-blog.com/migration-guide", "matchedWords": 42, "percentage": 22 }], "credits_used": 1 }

Result: Flagged — AI score 92%, plagiarism 27% duplicated.
Total credits used: 2.

Example 2: Verifying an original product review

An e-commerce site checks a customer-submitted product review:

AI Detection:
POST https://api.originality.ai/api/v1/scan/ai
Body: { "content": "I bought this blender last March after my old Vitamix finally died (RIP, 8 years of smoothies). The Ninja BN701 is louder than I expected — my cat literally bolts out of the kitchen — but it crushes frozen mango like nothing. The lid seal is a bit finicky, learned the hard way when I repainted my ceiling with acai. For $89 though, no complaints.", "aiModelVersion": "1", "storeScan": false }

Response:
{ "success": true, "score": { "ai": 0.04, "original": 0.96 }, "credits_used": 1 }

Plagiarism Check:
Response:
{ "success": true, "score": { "percentUnique": 100, "percentDuplicated": 0 }, "matches": [], "credits_used": 1 }

Result: Passes both checks — AI score 4%, 100% unique content.

Guidelines

Minimum 50 words required for AI detection
Best accuracy for English; other languages are supported but less reliable
Plagiarism check only works for publicly indexed web content
Heavily paraphrased AI text may evade detection
Does not detect AI in images, code, or structured data
Budget credits for bulk runs: estimate ~1 credit per 100 words per scan type
Handle HTTP 402 (insufficient credits) and 429 (rate limited) gracefully
Always pair automated detection with human review for final decisions