Configuration

new CodeIndexer(options)

const indexer = new CodeIndexer({
  // --- Required ---
  projectRoot: string,

  // --- Identity ---
  projectId?: string,           // stable ID for namespacing; default: sha1(projectRoot).slice(0,12)

  // --- Storage ---
  storageDir?: string,          // default: <projectRoot>/.lucerna

  // --- File selection ---
  include?: string[],           // default: **/* (all files — language detection filters unknowns)
  exclude?: string[],           // appended to defaults (node_modules, .git, dist, build, …)
                                // .gitignore at any depth is always applied automatically

  // --- Embeddings ---
  embeddingFunction?:
    | EmbeddingFunction         // custom instance
    | false,                    // disable semantic search, BM25 only
                                // undefined → auto: CloudflareEmbeddings if env vars set, else NomicCodeEmbeddings

  // --- Reranking ---
  rerankingFunction?:
    | RerankingFunction         // custom instance
    | false,                    // default: false (no reranking)

  // --- Chunking ---
  maxChunkTokens?: number,      // soft max tokens per chunk (1 token ≈ 4 chars); default: 1500

  // --- Watching ---
  watch?: boolean,              // start watching on initialize(); default: false
  watchDebounce?: number,       // debounce delay in ms; default: 500

  // --- Callbacks ---
  onIndexed?: (event: IndexEvent) => void,
  // IndexEvent: { type: 'indexed' | 'removed' | 'error', filePath, chunksAffected?, error? }
});

Embedding auto-detection

When embeddingFunction is not set, the indexer auto-selects:

1. CloudflareEmbeddings — if CLOUDFLARE_ACCOUNT_ID and CLOUDFLARE_API_TOKEN are set
2. NomicCodeEmbeddings — otherwise (local ONNX model, no API key required)

See Embeddings for all options.