n8n-io/n8n · error · NodeOperationError

Tool "${toolName}" was called but not found among connected

Error message

Tool "${toolName}" was called but not found among connected tools

What it means

Thrown during AlibabaCloud text/message tool-call processing when the LLM returns a function/tool call referencing a tool name that does not match any connected tool's name. The code iterates over connectedTools looking for a matching name; if none is found, toolResponse remains undefined and the error is thrown. This indicates a mismatch between what the model thinks is available and what the node actually has connected.

Source

Thrown at packages/@n8n/nodes-langchain/nodes/vendors/AlibabaCloud/actions/text/message.operation.ts:463

	for (const toolCall of toolCalls) {
		const toolName = toolCall.function.name;
		let toolInput: unknown;
		try {
			toolInput = JSON.parse(toolCall.function.arguments);
		} catch {
			toolInput = toolCall.function.arguments;
		}

		let toolResponse: unknown;
		for (const connectedTool of connectedTools) {
			if (connectedTool.name === toolName) {
				toolResponse = await connectedTool.invoke(toolInput);
			}
		}

		if (toolResponse === undefined) {
			throw new NodeOperationError(
				this.getNode(),
				`Tool "${toolName}" was called but not found among connected tools`,
			);
		}

		messages.push({
			role: 'tool',
			content:
				typeof toolResponse === 'object'
					? JSON.stringify(toolResponse)
					: String(toolResponse ?? ''),
			name: toolName,
		});
	}
}

View on GitHub (pinned to 5ac6606e81)

Solutions

1. Check which tools are connected to the AlibabaCloud node and verify their names match what the model is calling.
2. Inspect the tool call in the execution data — the error message includes the exact toolName the model tried to call.
3. Re-add the missing tool sub-node or rename the connected tool to match the model's expectation.
4. Restart the conversation/workflow execution so the model receives an updated tool list.

// Before processing tool calls, verify the called tool exists:
const connectedToolNames = connectedTools.map(t => t.name);
for (const toolCall of toolCalls ?? []) {
  if (!connectedToolNames.includes(toolCall.function.name)) {
    // log warning and skip, or provide a fallback response
    messages.push({
      role: 'tool',
      content: `Tool '${toolCall.function.name}' is not available.`,
      name: toolCall.function.name,
    });
  }
}

try {
  // ... tool call processing ...
} catch (error) {
  if (error instanceof NodeOperationError && error.message.includes('not found among connected tools')) {
    // provide a graceful fallback to the model
    messages.push({ role: 'tool', content: 'Tool not available, please use an alternative.', name: toolName });
  } else {
    throw error;
  }
}

When it happens

Trigger: The Alibaba Cloud Qwen model returns a tool_call with a function.name that doesn't match any connected LangChain tool's name. This happens when tools were connected/disconnected after the conversation context was established, when the model hallucinates a tool name, or when tool naming conventions differ between the model's training and the connected tools.

Common situations: A tool sub-node was removed or renamed after the workflow was last run; the model invents a tool name not in the actual tool list; tool names contain characters or casing that differs from what the model outputs; multiple tools with overlapping functionality cause the model to blend names.

Related errors

• LLM doesn't support binding tools
• Failed to parse output
• Run ${this.runId} is not suspended. Cannot resume.
• Reflector output must be valid JSON
• Reflector output must be a JSON object

AI-assisted analysis of n8n-io/n8n@5ac6606e81 (2026-08-12). Data as JSON: /api/errors/91d8c666da0cf378. Report an issue: GitHub.