Model Context Protocol (MCP) server for n8n workflow automation
Server has 11 tools with basic schemas and descriptions, but significant gaps prevent production use. All tools have descriptions (10-50 chars, below the 50-200 char baseline for LLM optimization), and input schemas are present but minimal. Most parameters lack descriptions entirely (e.g., 'nodes', 'connections', 'settings' in create_workflow have no guidance on structure or constraints). No output schemas documented. Error handling is generic ('Unknown tool', 'Unknown error') with no recovery guidance. Tool names follow verb_noun convention (good), but parameter naming is inconsistent (workflowId vs executionId mixing camelCase). No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite clear risk levels. Composition is reasonable (separate tools per action), but descriptions do not explain WHEN to use each tool or dependencies between them.
Activate a workflow
Create a new workflow in n8n
Deactivate a workflow
Delete an execution
Delete a workflow
Get a specific execution by ID
Get a specific workflow by ID
Complex parameters (nodes, connections, settings) lack any description of expected structure, format, or constraints. LLMs cannot infer whether 'nodes' is an array of objects, strings, or IDs, or what fields each node requires.
run_webhook has empty input schema (no properties defined) and vague description ('Run a webhook for workflow execution'). Unclear what parameters it accepts, what it does, or when to call it vs other execution tools.
No output schemas documented for any tool. LLMs cannot plan downstream calls or extract required fields (e.g., does get_workflow return workflow_id, name, nodes, status?). Breaks tool chaining.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 50 | 2026-07-28+ | v2 |
List all workflow executions
List all workflows from n8n
Run a webhook for workflow execution
Update an existing workflow
Tool descriptions are 20-50 characters, well below the 50-200 char baseline for LLM optimization. Descriptions lack WHEN to use, prerequisites, or state-modification warnings. E.g., 'Delete a workflow' does not warn that this is irreversible.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite clear risk levels (READ_ONLY, WRITE, DESTRUCTIVE). Agents cannot reason about safety or retry semantics without explicit hints.