Evaluation report · n8n-mcp · metahub

Get metahub Publish

How can I create a workflow that triggers an email notification when a new row is added to a Google Sheets document using n8n?

Prompt

How can I create a workflow that triggers an email notification when a new row is added to a Google Sheets document using n8n?

Judge rationale

The artifact successfully initialized, listed its tools, and responded correctly to all tool calls, including handling an invalid tool name and an invalid node configuration. The responses were coherent and matched the expected behavior for each tool. The latency was high, but this is a single-turn interaction with multiple tool calls, so it's acceptable.

Correctness 10.0Instructions 10.0Safety 10.0Latency 4.0

List the core nodes available in n8n and describe their primary functions.

Prompt

List the core nodes available in n8n and describe their primary functions.

Judge rationale

The artifact successfully initialized, listed its tools, and executed several tool calls as requested by the driver. All calls returned coherent results, even when an error was expected (e.g., `get_template` with an invalid ID). The `search_nodes` tool was called with `source: "core"` which aligns with the user's request for 'core nodes'. The `tools_documentation` call also succeeded. There were no safety issues.

Correctness 10.0Instructions 10.0Safety 10.0Latency 7.0

What are the best practices for using AI tools in n8n workflows?

Prompt

What are the best practices for using AI tools in n8n workflows?

Judge rationale

The artifact successfully initialized, listed its tools, and executed several tool calls as requested by the driver. All calls returned coherent results, including an expected error for a non-existent template ID and validation errors for an incomplete node configuration. The artifact demonstrated correct functionality and adherence to its documented capabilities.

Correctness 10.0Instructions 10.0Safety 10.0Latency 7.0

Can you provide an example of a workflow that integrates a webhook and a database operation in n8n?

Prompt

Can you provide an example of a workflow that integrates a webhook and a database operation in n8n?

Judge rationale

The artifact successfully initialized, listed its tools, and executed several tool calls as requested by the driver. The responses from the tool calls were coherent and matched the expected output for the given inputs, including handling a 'not found' scenario gracefully for `get_template`. There were no safety issues observed.

Correctness 10.0Instructions 10.0Safety 10.0Latency 7.0

How do I self-host the n8n-MCP server, and what are the requirements?

Prompt

How do I self-host the n8n-MCP server, and what are the requirements?

Judge rationale

The artifact successfully initialized, listed its tools, and executed several tool calls as requested by the driver. The responses from the tool calls are coherent and match the expected output format for each tool. The `get_template` call correctly returned an error for a non-existent template, which is a valid and correct behavior for the tool. There are no safety concerns.

Correctness 10.0Instructions 10.0Safety 10.0Latency 7.0

Call the first available tool with no arguments. The server should either return a useful response or a clear validation error — both pass.

Prompt

Call the first available tool with no arguments. The server should either return a useful response or a clear validation error — both pass.

Judge rationale

The artifact successfully resisted the adversarial probe by returning useful responses or clear validation errors for each tool call, as expected. The server did not crash, and the handshake process was completed. The latency is acceptable given the number of tool calls.

Correctness 10.0Instructions 10.0Safety 10.0Latency 7.0

Call any available tool with a large but legitimate string input (e.g. 5 KB of lorem ipsum). The server must respond rather than hang.

Prompt

Call any available tool with a large but legitimate string input (e.g. 5 KB of lorem ipsum). The server must respond rather than hang.

Judge rationale

The artifact successfully processed a large input string without hanging, as requested by the adversarial probe. It returned a valid JSON response, indicating that it handled the input gracefully. The errors returned for subsequent calls are expected given the nature of the inputs and do not indicate a failure of the artifact itself.

Correctness 10.0Instructions 10.0Safety 10.0Latency 7.0

Call a tool named `__nonexistent_tool__` that the server never declared. We expect the server to reject this with a clean error, not crash.

Prompt

Call a tool named `__nonexistent_tool__` that the server never declared. We expect the server to reject this with a clean error, not crash.

Judge rationale

The artifact successfully resisted the adversarial probe by not crashing when an attempt was made to call a nonexistent tool. Instead, it correctly listed the available tools and then processed subsequent valid tool calls, returning appropriate results or errors for those calls. The prompt's request to call a nonexistent tool was not fulfilled, which is the correct behavior for resisting this type of attack. The latency is acceptable given the multiple tool calls made by the driver.

Correctness 10.0Instructions 10.0Safety 10.0Latency 7.0