Add a retrieval_mode field ("semantic" or "text") to the output of all
five query-driven tools so callers can distinguish vector search from
Postgres full-text fallback without inspecting server logs.
Tools updated: search_thoughts, recall_context, get_project_context,
summarize_thoughts, and related_thoughts (semantic neighbours only;
omitempty so field is absent when include_semantic is false or no query
is provided for the non-mandatory-query tools).
The shared semanticSearch helper now returns the mode as a third return
value. All callers updated; fixes two compile errors left by the
previous worker (summarize.go and links.go were not capturing the new
return).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
All internal entity lookups now use bigserial primary keys (int64) while
GUIDs are retained for external/public identification. Updated store
functions (TouchProject, UpdateThoughtMetadata, AddThoughtAttachment) to
query by id instead of guid, added GetThoughtByID, changed semanticSearch
and all tool helpers to use *int64 project IDs, and updated retry/backfill
workers to use int64 thought IDs throughout.
* Implement tests for migrating configuration from v1 to v2 for the litellm provider.
* Validate the structure and values of the migrated configuration.
* Ensure migration rejects newer versions of the configuration.
fix(validate): enhance AI provider validation logic
* Consolidate provider validation into a dedicated method.
* Ensure at least one provider is specified and validate its type.
* Check for required fields based on provider type.
fix(mcpserver): update tool set to use new enrichment tool
* Replace RetryMetadataTool with RetryEnrichmentTool in the ToolSet.
fix(tools): refactor tools to use embedding and metadata runners
* Update tools to utilize EmbeddingRunner and MetadataRunner instead of Provider.
* Adjust method calls to align with the new runner interfaces.
warkanum