Search: Add ai_autocomplete_questions field for autocomplete

[reakaleek]

What

Add a new ai_autocomplete_questions AI enrichment field with a prompt targeting short, simple questions for search bar autocomplete. Keep the existing ai_questions field unchanged.

Why

The existing ai_questions generates detailed technical questions (6-15 words) useful for semantic search, but too complex for autocomplete suggestions. Users searching naturally type simpler queries like "What is agent builder?" rather than "How do I import external tools using Model Context Protocol?"

How

• Add AiAutocompleteQuestions property to DocumentationDocument with a prompt targeting 3-10 word questions
• Add lexical mapping with SearchAsYouType completion and Completion suggest multi-fields
• Add semantic text mapping for the new field
• Add suggest completion multi-field to existing ai_questions mapping (per Search: Add completion suggest field to ai_questions mapping #3108)

Test plan

• Verify build passes
• Review generated questions after next AI enrichment run to confirm they are simpler and more natural

🤖 Generated with Claude Code

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: 9bcdb1f6-9c88-404c-931e-ed7a818dbaf3

📥 Commits

Reviewing files that changed from the base of the PR and between 36841d3 and ae36a5a.

📒 Files selected for processing (1)

• src/Elastic.Documentation/Search/DocumentationMappingConfig.cs

🚧 Files skipped from review as they are similar to previous changes (1)

• src/Elastic.Documentation/Search/DocumentationMappingConfig.cs

---

📝 Walkthrough

Walkthrough

Added a new AI enrichment field AiAutocompleteQuestions to DocumentationDocument (type string[]?, JSON name ai_autocomplete_questions) decorated with [AiField(..., MinItems = 3, MaxItems = 5)], [Text], and [JsonIgnore(Condition = JsonIgnoreCondition.WhenWritingNull)]. Updated index mapping to include ai_autocomplete_questions in Lexical mappings with the synonyms_fixed_analyzer/synonyms_analyzer, plus two autocomplete-related multi-fields: a search_as_you_type subfield (with IndexOptions("offsets")) and a suggest completion subfield. Also added ai_autocomplete_questions.semantic_text to Semantic mappings. No other public signatures or runtime logic were changed.

Suggested labels

feature

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately and concisely summarizes the main change: adding the ai_autocomplete_questions field for autocomplete functionality.
Description check	✅ Passed	The description is well-structured with clear What/Why/How sections that directly relate to the changeset and explain the rationale for the new field.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

✨ Simplify code

• Create PR with simplified code
• Commit simplified code in branch south-booth

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against the current code and only fix it if needed.

Inline comments:
In `@src/Elastic.Documentation/Search/DocumentationDocument.cs`:
- Around line 150-151: The AiField attribute's prompt string on
DocumentationDocument is self-contradictory: it tells users to "Avoid jargon,
specific API names" but then includes the example "What is the bulk API?";
update the prompt in the AiField attribute to be consistent by either removing
the rule about avoiding specific API names or replacing the example with a
non-API-specific example (e.g., "What is bulk indexing?") so the guidance and
examples match; locate the AiField attribute text on the property in
DocumentationDocument and make the prompt and examples consistent.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro Plus

Run ID: 92e1a1a3-4c37-4787-9226-32bf2338991b

📥 Commits

Reviewing files that changed from the base of the PR and between 78603a9 and 39349a6.

📒 Files selected for processing (1)

• src/Elastic.Documentation/Search/DocumentationDocument.cs

[Mpdreamz]

Should we have both? I feel the current will become super useful when we do more data/ai generated curation in the future.