ruvnet/ruflo · error · Error

each record requires a non-empty numeric vector

Error message

each record requires a non-empty numeric vector

What it means

Thrown by the agenticow_ingest handler (agenticow-tools.ts:132) when any record's vector field is not a non-empty array. agenticow stores pre-computed numeric vectors (HNSW-indexed) — text is optional metadata, not something that gets embedded on the fly — so every record must carry its own non-empty vector.

Solutions

1. Pre-compute embeddings and pass a real number[] per record (e.g. via the claude-flow embeddings tools) — dimension must match the file's dimension
2. Parse stringified vectors before sending: JSON.parse(rec.vector)
3. Filter or reject records lacking vectors before the call, and align the field name to 'vector'

Example fix

// before
await callTool('agenticow_ingest', {
  path: p, dimension: 384,
  records: [{ text: 'a doc with no embedding' }], // throws
});

// after
const vectors = await embed(texts); // pre-compute
await callTool('agenticow_ingest', {
  path: p, dimension: 384,
  records: texts.map((t, i) => ({ vector: vectors[i], text: t })),
});

function isNumericVector(v: unknown): v is number[] {
  return Array.isArray(v) && v.length > 0 && v.every((n) => typeof n === 'number' && Number.isFinite(n));
}
const bad = records.filter((r) => !isNumericVector(r.vector));
if (bad.length) throw new TypeError(`${bad.length} records lack a numeric vector`);

function isNumericVector(v: unknown): v is number[] {
  return Array.isArray(v) && v.length > 0 && v.every((n) => typeof n === 'number' && Number.isFinite(n));
}

When it happens

Trigger: A record with vector omitted (e.g. {text: 'doc'} expecting auto-embedding), vector: [], vector passed as a string ('[0.1,0.2]') or comma-separated '0.1,0.2', or vector: null. One bad record in the batch fails the entire call.

Common situations: Assuming the tool embeds text like a RAG pipeline; vectors JSON.stringify'd twice so they arrive as strings; sparse sources where some rows lack embeddings; undefined after destructuring a renamed field (e.g. record.embedding vs record.vector).

Related errors

• vector must be a non-empty numeric array
• at least one candidate is required
• candidate must ingest at least one vector
• label exceeds 256 chars
• label is required

AI-assisted analysis of ruvnet/ruflo@fa13ee4ad6 (2026-08-18). Data as JSON: /api/errors/69b6d03db9508470. Report an issue: GitHub.

Appendix: source

Thrown at v3/@claude-flow/cli/src/mcp-tools/agenticow-tools.ts:132

            required: ['vector'],
          },
        },
        dimension: { type: 'integer', description: 'Vector dimension (required only when path does not exist yet)' },
      },
      required: ['path', 'records'],
    },
    handler: async (input) => {
      const api = await loadAgenticow();
      if (!api) return degradedResult('agenticow-not-found');

      const path = resolveMemoryPath(String(input.path));
      const records = input.records as Array<{ id?: number; vector: number[]; text?: string }>;
      if (!Array.isArray(records) || records.length === 0) {
        throw new Error('records must be a non-empty array of {id?, vector, text?}');
      }
      for (const r of records) {
        if (!Array.isArray(r.vector) || r.vector.length === 0) {
          throw new Error('each record requires a non-empty numeric vector');
        }
      }
      const dim = (input.dimension as number | undefined) ?? records[0].vector.length;
      const mem = await openWithLineage(api, path, dim);
      try {
        const result = await mem.ingest(records.map((r) => ({
          ...(typeof r.id === 'number' ? { id: r.id } : {}),
          vector: r.vector,
          ...(r.text !== undefined ? { text: r.text } : {}),
        })));
        await mem.save?.(manifestFor(path));
        return { success: true, path, ingested: result };
      } finally {
        await mem.close?.();
      }
    },
  },
  {

View on GitHub (pinned to fa13ee4ad6)