langflow-ai/langflow · error · Error

Invalid flow data

Error message

Invalid flow data

What it means

Thrown by the useUploadFlow hook when a parsed upload file yields a flow object that has no `data` property. Langflow expects every uploaded JSON entry to be a flow object whose graph lives under `data` (nodes/edges); an entry without it is treated as structurally invalid and the whole upload is aborted before any flow is saved or pasted.

Source

Thrown at src/frontend/src/hooks/flows/use-upload-flow.ts:88

      ) {
        throw new Error(
          "You cannot upload a component as a flow or vice versa",
        );
      } else {
        let currentPosition = position;
        for (const flow of flows) {
          if (flow.data) {
            if (currentPosition) {
              paste(flow.data, currentPosition);
              currentPosition = {
                x: currentPosition.x + 50,
                y: currentPosition.y + 50,
              };
            } else {
              await addFlow({ flow });
            }
          } else {
            throw new Error("Invalid flow data");
          }
        }
      }
    } catch (e) {
      throw e;
    }
  };

  return uploadFlow;
};

export default useUploadFlow;

View on GitHub (pinned to 976ec789d2)

Solutions

1. Open the uploaded JSON and confirm each top-level object contains a `data` object with `nodes` and `edges`
2. Re-export the flow from Langflow's UI (Flow Settings > Export) rather than copying a flow-list API response
3. If importing multiple flows, ensure the file is an array where EVERY element has `data` — one bad entry aborts the loop
4. Validate the file shape client-side before calling uploadFlow (see type guard below)

Example fix

// before
await uploadFlow({ files }); // throws mid-loop on first entry without data

// after
const flows = JSON.parse(await files[0].text());
const ok = (Array.isArray(flows) ? flows : [flows]).every(
  (f) => f && typeof f === "object" && f.data && Array.isArray(f.data.nodes),
);
if (!ok) throw new Error("File is not a valid Langflow flow export");
await uploadFlow({ files });

const isFlowExport = (v: unknown): v is { data: { nodes: unknown[]; edges: unknown[] } } =>
  typeof v === "object" && v !== null &&
  "data" in v && typeof (v as any).data === "object" &&
  Array.isArray((v as any).data?.nodes);

const parsed = JSON.parse(text);
const flows = Array.isArray(parsed) ? parsed : [parsed];
if (!flows.every(isFlowExport)) {
  alert("File is not a valid Langflow flow export (missing data.nodes)");
}

function isFlowExport(v: unknown): v is { data: { nodes: unknown[]; edges: unknown[] } } {
  return (
    typeof v === "object" && v !== null &&
    "data" in v && typeof (v as Record<string, unknown>).data === "object"
  );
}

try {
  await uploadFlow({ files });
} catch (e) {
  if (e instanceof Error && e.message === "Invalid flow data") {
    // show a file-format message; do not retry the same file
  } else throw e;
}

When it happens

Trigger: Uploading a JSON file via the flow-import UI (or drag-drop onto the canvas with a paste position) where getFlowsFromFiles returns objects lacking a `data` key — e.g. a hand-written JSON, an API response wrapper like {id, name} without the graph, or a partially exported/corrupted flow file.

Common situations: Users exporting from a different Langflow version or another tool, editing exported JSON and accidentally deleting the `data` field, or uploading a file that contains flow metadata (list endpoint output) instead of the full flow export.

Related errors

• Invalid JSON format.
• No valid MCP server found in the input.
• Deployment name is required
• error: invalid JSON in {source}: {exc}
• error: {source} entries field must be a list

AI-assisted analysis of langflow-ai/langflow@976ec789d2 (2026-08-14). Data as JSON: /api/errors/2b3f574f2e044f88. Report an issue: GitHub.