AI That Knows Your Content Model. Not Just Your Text

Most CMS AI features are wrappers. You get a text box that calls GPT, a button that rewrites whatever you've selected, and a marketing page that says "AI-powered." What you don't get is an AI that understands the difference between a hero_title field and a body_text field. Or one that knows your content has 12 locales and 4 of them are missing meta descriptions.

That's not integration. That's decoration.

b10cks' AI works with your content model, not around it.

SEO metadata is tedious to write, easy to forget, and critical to get right. b10cks AI reads your published content and generates meta titles and descriptions that actually match what's on the page, not generic filler.

How it works:

• Select any content entry: page, article, product, whatever
• Trigger meta tag generation from the editor panel
• AI reads the content and writes tags calibrated to title length, description length, and keyword relevance
• Review and edit before saving – you're always in control

What changes:

• Every piece of content gets a meta title and description that reflects its actual text
• Content teams don't need SEO training to produce search-ready pages
• No more copy-paste from the body text and calling it a description

Meta tags are included in every plan. No SEO add-on. No extra module.

Your content library has structure: articles, pages, products, locales, statuses, versions. b10cks AI can read that structure and answer questions about it in plain language.

Things you can ask:

• "Which articles are missing a meta description?"
• "Show me all content items that are in draft state in the German locale"
• "How many pages don't have a featured image set?"
• "Which blog posts were last updated more than 6 months ago?"

And write to it:

• "Set the status of all draft articles in the en-GB locale to ready for review"
• "Add the tag 'seasonal' to all content items published in December"
• "Write me a typical page tree for an SaaS Startup"
• "Generate meta descriptions for all pages in the /products/ section that don't have one"

This is AI working inside your content infrastructure not just responding to a prompt in a floating widget.

Localization is one of the most time-consuming parts of managing multilingual content. b10cks AI makes it immediate.

How translation works:

• Open any content entry
• Switch to a target locale that's missing content
• Trigger AI translation
• Watch the translation stream field by field: headline, body, alt text, meta description, all of it
• Review, adjust, save

What makes it different:

Translation in b10cks is context-aware. The AI sees:

• Field types (it knows a slug field shouldn't be translated, a meta_description field should be concise)
• Existing translations in other locales as reference
• Your block schema – so it understands what each field is for

You're not translating a blob of text. You're translating structured content, and the AI knows the difference.

XLIFF compatible: If you work with professional translation agencies, b10cks also supports XLIFF export and import – use AI translation for speed, human review for precision, or combine both in the same workflow.

An asset library grows fast. Images, videos, documents, graphics – hundreds of files that need titles, descriptions, and alt text to be findable and accessible. Writing that metadata manually is exactly the kind of work that gets skipped.

b10cks AI classifies your assets automatically:

For images:

• Generates a descriptive title based on visual content
• Writes an alt text description suitable for screen readers and SEO
• Suggests tags based on subject matter, style, and context

For documents:

• Generates a title and summary from the document content
• Extracts key topics for tag suggestions

How to use it:

• Select one asset or a batch of assets
• Trigger classification from the asset library
• Review the generated metadata – it's a starting point, not a final answer
• Approve individually or in bulk

Your asset library becomes searchable and accessible without a metadata sprint every time you upload a batch of new files.

Open the AI dock while editing any content item and you have a writing assistant that knows exactly what you're building:

• Write from scratch – describe what the field should contain and let AI draft it
• Rewrite and refine – paste existing content and ask AI to change the tone, shorten it, or make it more direct
• Summarize – generate a summary of a long article for use in a teaser or description field
• Expand – turn a bullet list of notes into a full paragraph

The AI sees your block schema. It knows whether it's writing a hero headline (short, punchy) or a product description (detailed, keyword-aware). You don't have to explain the context – it's already there.