How it works
Your organization has a primary language and can enable additional languages. When you run enrichment, Cernel generates content in each enabled language. Each language can carry its own translation prompt: instructions Cernel applies whenever it translates content into that language, like grammar rules, cultural preferences, or terminology that should be used. Content generated directly in a language follows the language’s generation prompt instead: the instruction Cernel adds to your agents whenever they write content in that language. Languages are managed under Tools > Languages & markets:- The languages list is where you enable or disable languages and add new ones.
- Each language has its own detail page - open it to edit that language’s generation prompt and translation prompt, and choose whether content for it is always written in that language.
Every language is also a market
Every language you add is also a market. Cernel writes your product content in the language, and Search demand reads what shoppers search for in that market. Choose the code that matches where you sell, and include the country when it matters:en_GB and en_US are the same language but two markets, each with its own search demand.
Why you would use this
Reach customers in their language
Language-appropriate content
Setting up languages
Enable languages

Add a new language
en for English, da for Danish, or a region-specific code like sv_FI for Swedish (Finland), and click Add language & market. Once added, the language appears in your table, and it is also a market that Search demand can read.
Set a language's generation prompt
- Clear the box and click Save to go back to Cernel’s standard prompt.
- Turn Always output in [language] off to add no language instruction at all, and control the output language entirely through your own agent prompts. The generation prompt can’t be edited while the toggle is off.
- The toggle is saved together with the prompt, so click Save after switching it.

A language's Generate tab, with its generation prompt
Configure a language's translation prompt
- Grammar and style rules specific to the language
- Cultural conventions and preferences
- Terminology that should (or shouldn’t) be used
/ in the prompt editor to pull in a Data Source for reference, then click Save.Cernel applies the translation prompt whenever it translates a Text or HTML value into this language, for example when the language is set to Translate in an enrichment run. When the language is set to Generate, the agent writes the value from its own prompt plus the language’s generation prompt, and the translation prompt isn’t used. To give an agent language-specific rules for generated content, use its Languages tab.
A language's Translate tab, with its translation prompt
Fine-tuning language output
A language’s generation and translation prompts set how content is written and translated in that language across your whole organization. Two further controls let you refine language behavior more precisely: per agent, and per attribute.Give an agent its own language instructions
Text and HTML agents can carry their own instructions for each language, on top of the organization-wide settings. Open a Text or HTML agent in the Agent Sandbox, go to the Languages tab, and you’ll find one box per enabled language. Whatever you write in a box is applied when the agent writes a value in that language, and when a value is translated into it. Leave a box empty to add nothing for that language. Type/ in a box to reference a Data Source.
Use this when a single agent needs a language rule the others don’t, for example a size-guide agent that must use metric phrasing in German, without changing how any other agent writes German.

Choose which languages an attribute is written in
By default, each enrichment run decides which languages an attribute is written in, one plan for the whole run. You can instead fix that choice for a single attribute. Open the attribute’s agent settings, expand Enrichment Settings, and turn on Override language settings. A per-language plan appears where each enabled language is set to Generate, Translate, or Skip, the same editor you use when starting a manual enrichment. Once it’s on, that plan replaces whatever a run asks for, but only for this attribute. A new plan starts the way a run does: the attribute is written in your primary language and translated into the rest. Because the plan is a fixed list rather than a rule, a language you enable later sits outside it and is skipped for this attribute until you add it. Cernel notes under the editor which languages are currently left out, so the gap is easy to spot.Advanced configuration
Global Prompts for attribute types
Global Prompts for attribute types
Translations for select attribute values
Translations for select attribute values
Disabling a language
Disabling a language
Frequently asked questions
Do I need separate agents for each language?
Do I need separate agents for each language?
Can I enrich in only one language at a time?
Can I enrich in only one language at a time?
Which AI provider translates my content?
Which AI provider translates my content?
What's the difference between a translation prompt and an agent prompt?
What's the difference between a translation prompt and an agent prompt?
Can I change how Cernel tells agents to write in a language?
Can I change how Cernel tells agents to write in a language?
How do an agent's language instructions relate to a language's translation prompt?
How do an agent's language instructions relate to a language's translation prompt?
How do I give each language its own reference content, like a per-market keyword list?
How do I give each language its own reference content, like a per-market keyword list?
seo/keywords-es, seo/keywords-fr), then open the agent, go to its Languages tab, and in each language’s box type / to reference only that market’s entry: the Spanish entry in the Spanish box, the French entry in the French box, and so on. A run in a given language then only pulls in the content from that language’s box, so each market sees only its own keywords. You don’t need separate agents to do this. See Give an agent its own language instructions.When a language is set to Translate, does it keep the original prompt's rules?
When a language is set to Translate, does it keep the original prompt's rules?
