Copyleaks
| 开发者 | copyleaks |
|---|---|
| 更新时间 | 2026年8月26日 19:29 |
| PHP版本: | 7.4 及以上 |
| WordPress版本: | 7.0 |
| 版权: | GPL-2.0-or-later |
| 版权网址: | 版权信息 |
详情介绍:
- AI Detection Score — an AI-generated content score for every scanned post or page, shown directly in the posts list.
- Plagiarism Score — a plagiarism / matched-content score for every scanned post or page.
- Grammar Suggestions — a grammar corrections count for scanned content.
- Auto-Scans — automatic scanning when a post or page reaches a chosen status (Pending Review or Published), plus manual on-demand scanning and a retry button for failed scans.
- Encryption and Data Security — API credentials are stored encrypted, and content is transmitted securely to the Copyleaks API.
安装:
- In your WordPress admin, go to Plugins → Add Plugin.
- Search for Copyleaks.
- Click Install Now, then Activate.
- Download the plugin ZIP and upload it via Plugins → Add Plugin → Upload Plugin, or extract the
copyleaksfolder into/wp-content/plugins/. - Activate the plugin through the Plugins screen.
- Go to Settings → Copyleaks and enter your API Key and Secret obtained from your Copyleaks account.
- Click Register to connect your site.
- In the Configuration tab, choose which post status triggers an automatic scan (Pending Review, Published, or Manual).
屏幕截图:
常见问题:
Do I need a Copyleaks account?
Yes. Copyleaks is a paid, credit-based service, so an active Copyleaks account with API access is required and each scan consumes credits from your account. Sign up and view pricing at https://copyleaks.com.
What data is sent to Copyleaks?
The plain-text content of posts and pages (HTML stripped), the post title, and the post author's WordPress username, email address, and user ID are submitted to the Copyleaks API for analysis. No site-visitor data is transmitted.
Is the API Secret stored securely?
Yes. The API Secret is encrypted before being stored in the WordPress database.
Can I scan posts manually?
Yes. Set the scan trigger to Manual in the Configuration tab. A "Scan now" button will appear in the posts list for any post that has not yet been scanned.
How does AI detection work?
When a Large Language Model writes a sentence, it uses all its pre-training data to create a response or an output based on statistical modeling. AI detectors can recognize this pattern and flag certain content as AI-generated because it significantly differs from the way humans write. AI detectors can be trained on human-written content samples, so the presence of an AI text generator becomes clearer compared to human writing.
What models can the AI Detector detect?
Copyleaks can detect content from GPT-5, GPT-4, ChatGPT, Claude, Gemini, DeepSeek, Llama (Meta), Bloom, Rytr, Jasper, and more.
Why is there a minimum and maximum text requirement for some AI Detector content scans?
The Copyleaks detection models need a certain volume of text to determine AI presence accurately. More characters improve detection confidence.
- Minimum: 255 characters
- Maximum: 2,000 pages (no character maximum)
What data protection does Copyleaks have?
Copyleaks prioritizes data privacy and compliance. Current certifications include:
- SOC2 & SOC3
- GDPR
- PCI DSS
- NIST RMF
Does the Copyleaks AI Detector flag writing tools like Grammarly as AI?
Not all writing tool features are flagged. Grammarly's AI-based rewriting features may trigger detection, while basic grammar and spell-check functions typically do not. The AI Detector is designed to distinguish between minor edits and generative rewriting, helping ensure fair and accurate analysis.
What happens to my data when I uninstall the plugin?
All plugin data is removed on uninstall: scan records, options, transients, and the custom database table are all deleted.
更新日志:
- Initial release.