內容簡介
OneClickContent - Image Detail Generator 是一款免費的 WordPress 外掛，透過使用者自有的 OpenAI 或 Gemini API 金鑰，自動為媒體庫中的圖片生成 AI 產生的元資料，提升 SEO 效果並改善無障礙性。
【主要功能】
• 自動生成圖片的 alt text、標題、說明與標題文字
• 支援批量生成媒體庫中的圖片元資料
• 可手動編輯生成的元資料
• 提供多語言輸出選項
• 具備預檢計數與乾跑功能
• 控制更新的欄位與恢復變更歷史
外掛標籤
開發者團隊
② 後台搜尋「OneClickContent – Image Detail Generator」→ 直接安裝（推薦）
原文外掛簡介
OneClickContent – Image Detail Generator is a free, bring-your-own-key plugin for enhancing existing WordPress images with AI-generated metadata.
Use your own OpenAI or Gemini API key to generate:
– alt text
– image titles
– captions
– descriptions
This plugin is focused on metadata generation for images already in your Media Library. It does not create images.
SVG attachments remain available for audits and manual editing. AI generation skips them without making a provider request.
Key Benefits:
– Improve SEO for image search and content relevance
– Improve accessibility with descriptive alt text
– Save time on repetitive metadata writing
– Use your own provider account and models
– Bulk-generate metadata across your library
– Control which fields get updated
– Run preflight counts and dry runs before making provider requests
– Review, edit, approve, reject, or defer field-level suggestions
– Process resumable batches with pause, resume, cancel, and retry controls
– Restore individual fields or complete batches from immutable change history
Features
Bring-your-own-key support for OpenAI and Gemini
Single-image generation inside the Media Library
Bulk generation for existing image attachments
Auto-generate on upload
Multilingual output based on plugin settings
Manual editing after generation
Forward-compatible OpenAI model filtering for GPT-5.6 API model IDs when OpenAI exposes them to API accounts
Safe fill-missing, suggestion, overwrite-confirmation, and dry-run modes
Confidence-aware review with decorative-image decisions
Custom capabilities, CSV exports, environment-managed API keys, and WP-CLI commands
External Services
This plugin connects directly to the AI provider you configure.
OpenAI
Model-list endpoint: https://api.openai.com/v1/models
When used: when an administrator validates an OpenAI API key and loads compatible model choices
Data sent for model listing: the configured OpenAI API key in the authorization header; no image or site content is sent
Generation endpoint: https://api.openai.com/v1/chat/completions
When used: when an authorized user requests image metadata generation, including upload processing, single-image generation, or a batch
Data sent for generation: the configured OpenAI API key, image data, filename, existing title/description/alt text/caption, site name, configured organization and editorial guidance, and the parent content title/excerpt/type when available. Published parent context is included by default; draft or private parent context is included only when the corresponding administrator opt-in is enabled.
Purpose: validate credentials, provide compatible model choices, and generate title, description, alt text, and caption
Terms: https://openai.com/policies/terms-of-use/
Privacy: https://openai.com/policies/privacy-policy/
Gemini
Model-list endpoint: https://generativelanguage.googleapis.com/v1beta/models?key={api_key}
When used: when an administrator validates a Gemini API key and loads compatible model choices
Data sent for model listing: the configured Gemini API key; no image or site content is sent
Generation endpoint: https://generativelanguage.googleapis.com/v1beta/models/{model}:generateContent?key={api_key}
When used: when an authorized user requests image metadata generation, including upload processing, single-image generation, or a batch
Data sent for generation: the configured Gemini API key, image data, filename, existing title/description/alt text/caption, site name, configured organization and editorial guidance, and the parent content title/excerpt/type when available. Published parent context is included by default; draft or private parent context is included only when the corresponding administrator opt-in is enabled.
Purpose: validate credentials, provide compatible model choices, and generate title, description, alt text, and caption
Terms: https://ai.google.dev/terms
Privacy: https://policies.google.com/privacy
Third-party Libraries
DataTables (with Buttons and HTML5 Export): (v2.2.2, Buttons v3.2.2) used for table display and export functionality.
Sources:
DataTables GitHub
Buttons GitHub
Source Code
Full source code and unminified JavaScript are available on GitHub.
Quick Start
Keep overwrite disabled for the safest starting point.
Open Image Metadata > Image Library.
Find an image and click Preview.
Compare the current and suggested values. Preview does not change the image.
Edit a suggestion if needed, then click Use this suggestion for each value you approve.
Use Generate when you want the current plugin rules applied automatically. With the recommended defaults, Generate fills empty selected fields and preserves existing metadata.
Whitespace-only fields are treated as empty. SVG attachments can be edited manually, but AI generation skips them.
How to Process a Batch
Open Image Metadata > Dashboard and review what is missing.
Under Choose what to run, select the fields and a small starting batch such as 10 or 25 images.
Leave the mode on Preview only – change nothing, then click Run Preview.
Review the estimated images, fields, provider requests, and cost.
Open Advanced batch options and choose one of these modes:
Create suggestions for review saves proposed values without changing attachments.
Fill missing fields applies values only where selected fields are empty.
Overwrite selected fields replaces existing values only when overwrite is enabled and immediately confirmed.
Run the batch, then open Image Metadata > Batches to monitor it.
Batches run in the background. The Batches screen provides pause, resume, cancel, and retry controls, so the original browser page does not need to remain open.
Your field, batch-size, mode, ordering, and missing-field choices are remembered when you run the form. Overwrite confirmation is never remembered.
How to Review and Restore
Open Image Metadata > Image Library and filter for images with suggestions ready. Each suggestion can be approved, edited before approval, rejected, or left for later.
The plugin records metadata before applying a workflow change:
Open Image Metadata > History to restore an individual field.
Open Image Metadata > Batches to restore a completed batch.
A normal restore protects values that someone edited after the recorded plugin change. An individual field can be force-restored only after explicit confirmation.
Recommended First Run
Enable only Alternative text and Title.
Keep overwrite disabled.
Preview 10 images.
Check several results against the images and page context.
Run Fill missing fields on a small batch.
Confirm the results in the Image Library before increasing the batch size.
Advanced workflow settings contain processing limits, request ceilings, retry behavior, editorial guidance, context controls, overwrite permissions, history retention, reports, capabilities, and WP-CLI support. Most sites can leave these settings at their defaults.
