mastra-ai/mastra · error

No search configuration available. Provide bm25 or vector co

Error message

No search configuration available. Provide bm25 or vector config.

What it means

The engine was asked to search with no mode specified, so it tried to auto-select one, but neither vector nor BM25 was configured at all. With no index and no embedding pipeline there is nothing to search against, so it throws rather than returning empty results.

Source

Thrown at packages/core/src/workspace/search/search-engine.ts:672

      }
      if (requestedMode === 'hybrid' && !this.canHybrid) {
        throw new Error('Hybrid search requires both vector and BM25 configuration.');
      }
      return requestedMode;
    }

    // Auto-determine based on available configuration
    if (this.canHybrid) {
      return 'hybrid';
    }
    if (this.canVector) {
      return 'vector';
    }
    if (this.canBM25) {
      return 'bm25';
    }

    throw new Error('No search configuration available. Provide bm25 or vector config.');
  }

  /**
   * Embed a single text, dispatching to the batch path with a one-element array
   * when the configured embedder is batch-capable.
   */
  async #embedOne(text: string): Promise<number[]> {
    if (!this.#vectorConfig) {
      throw new Error('Vector configuration is required to embed text.');
    }
    const { embedder } = this.#vectorConfig;
    if (isBatchEmbedder(embedder)) {
      const [embedding] = await embedder([text]);
      if (!embedding) {
        throw new Error('Batch embedder returned no embedding for input text.');
      }
      return embedding;
    }

View on GitHub (pinned to 75dd419e61)

Solutions

1. Construct the SearchEngine with at least one backend: a `bm25` config for keyword search or a `vectorConfig` (vectorStore + embedder + indexName) for semantic search.
2. If search should be optional, check `engine.canBM25 || engine.canVector` before calling search and skip/handle the no-search case.
3. Verify your config/env loading actually reaches the SearchEngine constructor (log the options before constructing).

Example fix

// before
const engine = new SearchEngine();
await engine.search('query'); // throws

// after
const engine = new SearchEngine({
  bm25: {},
  vectorConfig: { vectorStore, embedder, indexName: 'docs' },
});
await engine.search('query');

if (!engine.canBM25 && !engine.canVector) {
  throw new Error('SearchEngine has no search backend configured; provide bm25 or vectorConfig.');
}
return engine.search(query);

try {
  results = await engine.search(query);
} catch (e) {
  if (e instanceof Error && e.message.includes('No search configuration available')) {
    results = []; // or configure a backend and retry
  } else {
    throw e;
  }
}

When it happens

Trigger: Calling `search(query)` (or `search(query, {})`) on a SearchEngine constructed with neither `vectorConfig` nor a BM25 config — `canVector`, `canBM25`, and `canHybrid` are all false.

Common situations: Instantiating SearchEngine with an empty/omitted options object in a Workspace setup; config loading that silently drops the search options (bad env vars, missing embedder key); code that defers configuring search but already calls search during startup or tests.

Related errors

• BM25 search requires BM25 configuration.
• Hybrid search requires both vector and BM25 configuration.
• Vector search requires vector configuration.
• Vector configuration is required to embed text.
• Workspace does not have search configured

AI-assisted analysis of mastra-ai/mastra@75dd419e61 (2026-08-30). Data as JSON: /api/errors/1bb74cd7b8523e6f. Report an issue: GitHub.