ruvnet/ruflo · error · Error

Failed to create share link

Error message

Failed to create share link

What it means

task_assign (simple implementation path) looks the task up in the in-memory taskStore and throws 'Task not found: <id>' when the Map has no entry. This runs after an orchestrator attempt failed ('Failed to assign task via orchestrator:' on console). Note the ordering of guards: an already-assigned task only soft-returns assigned:false when reassign is not set; a missing task throws before any of those checks.

Solutions

1. Ensure task_create and task_assign run against the same MCP server process; re-create the task after any restart and use the new ID
2. List tasks before assigning to confirm the ID is live
3. Fix the orchestrator if console logs show the assign attempt failing there — the local fallback only knows local tasks
4. Keep a durable taskId registry in your orchestration layer so restarts trigger re-creation instead of dead-ID assignments

Example fix

// before
await client.callTool('task_assign', { taskId: oldTaskId, agentId: 'worker-1' }); // throws [1135]

// after
const { task } = await client.callTool('task_create', { type: 'implementation', description: '...' });
await client.callTool('task_assign', { taskId: task.id, agentId: 'worker-1' });

const { tasks } = await client.callTool('task_list', {});
const assignable = tasks.find(t => t.id === taskId && !['completed', 'cancelled', 'failed'].includes(t.status));
if (!assignable) throw new Error(`no assignable task ${taskId}`);

try {
  await client.callTool('task_assign', { taskId, agentId });
} catch (e) {
  if (e instanceof Error && e.message.startsWith('Task not found')) {
    const { task } = await client.callTool('task_create', { type: 'implementation', description: `recreated ${taskId}` });
    return client.callTool('task_assign', { taskId: task.id, agentId });
  }
  throw e;
}

When it happens

Trigger: task_assign for an ID created in a previous process; orchestrator down or throwing so the code falls through to an empty local store; assigning to a task ID that was pruned or never created; passing an agent spawn name as taskId by mistake.

Common situations: Restarting the MCP server between creating tasks and dispatching them to agents; multi-instance deployments where task creation and assignment hit different server processes; scripts reusing IDs from a prior campaign.

Related errors

• JSON.stringify(response.error)
• SSRF guard: private/loopback host rejected
• Task not found
• User not found
• User token not found

AI-assisted analysis of ruvnet/ruflo@fa13ee4ad6 (2026-08-18). Data as JSON: /api/errors/8fb8055856bc31c7. Report an issue: GitHub.

Appendix: source

Thrown at ruflo/src/ruvocal/src/lib/createShareLink.ts:22

// Returns a public share URL for a conversation id.
// If `id` is already a 7-char share id, no network call is made.
export async function createShareLink(id: string): Promise<string> {
	const prefix =
		page.data.publicConfig.PUBLIC_SHARE_PREFIX ||
		`${page.data.publicConfig.PUBLIC_ORIGIN || page.url.origin}${base}`;

	if (id.length === 7) {
		return `${prefix}/r/${id}`;
	}

	const res = await fetch(`${base}/conversation/${id}/share`, {
		method: "POST",
		headers: { "Content-Type": "application/json" },
	});

	if (!res.ok) {
		const text = await res.text().catch(() => "");
		throw new Error(text || "Failed to create share link");
	}

	const { shareId } = await res.json();
	return `${prefix}/r/${shareId}`;
}

View on GitHub (pinned to fa13ee4ad6)