Reindex Search Index

Refresh the search projection from authoritative storage.

Most calls perform a full semantic rebuild of:

• All memories (with search embeddings)
• All thread messages
• All Library sources and source chunks
• All communities
• All entities

If the only pending marker is the 0.8.4 Content Store thread-message migration and semantic index metadata already matches the active embedding backend, this endpoint only copies/verifies messages, rebuilds the FTS-only messages_index, and skips embedding initialization.

The embedding model is platform-specific:

• macOS Apple Silicon: Qwen3-Embedding via mlx-embeddings
• non-Apple-Silicon (Windows/Linux/macOS Intel): BGE-M3 via FastEmbed/ONNX

This is a heavy operation and should only be triggered:

• After first downloading the search embedding model
• After a data migration
• When explicitly requested by the user

curl -X POST "https://example.com/search-index/reindex"

{
  "success": true,
  "memories": 0,
  "messages": 0,
  "sources": 0,
  "source_chunks": 0,
  "communities": 0,
  "entities": 0,
  "errors": [
    "string"
  ],
  "message": "",
  "restart_recommended": false
}