Content
56%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides comprehensive, actionable guidance for n8n-mcp tools with good workflow sequences and validation patterns. Its main weakness is significant verbosity—information is repeated across multiple sections (Quick Reference, Tool Selection Guide, Tool Usage Patterns, Unified Tool Reference, Summary all cover overlapping content), and timing statistics add little value. The content would benefit from aggressive deduplication and pushing detailed reference material into the referenced sub-guides.
Suggestions
Consolidate overlapping sections: merge Quick Reference, Tool Selection Guide, Tool Usage Patterns, and Unified Tool Reference into a single concise reference, moving detailed examples to the referenced sub-guides (SEARCH_GUIDE.md, VALIDATION_GUIDE.md, WORKFLOW_GUIDE.md).
Remove timing statistics (18s average, 56s between edits, 23s thinking) as they don't help Claude use the tools and add unnecessary tokens.
Cut the Summary section's repetition of points already made in the body—a brief 'Common Workflow' sequence and links to sub-guides would suffice.
Move the detailed Template Usage and Self-Help Tools sections to their own guide files, keeping only a brief mention and link in the main SKILL.md.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is significantly verbose at ~400+ lines. It repeats information across sections (e.g., nodeType formats explained in Critical section, Common Mistakes, and Summary), includes timing statistics Claude doesn't need (18s average, 56s between edits), and has redundant examples. The Quick Reference table, Tool Selection Guide, Tool Usage Patterns, and Unified Tool Reference all cover overlapping ground. Many sections explain things Claude would understand from a single example. | 2 / 5 |
Actionability | The skill provides concrete, executable code examples with specific tool calls, parameter formats, and expected return values. The nodeType format distinction is clearly illustrated with correct/incorrect examples. However, the JavaScript examples use `await` syntax that isn't directly copy-pasteable as MCP tool calls—they're illustrative rather than truly executable in the MCP context. The validation loop and workflow editing patterns are concrete and useful. | 4 / 5 |
Workflow Clarity | Multi-step workflows are clearly sequenced (search → get_node → validate → create → validate → update → activate) with validation checkpoints explicitly included. The validation loop pattern includes a fix-and-retry feedback loop. However, the workflow management section could be clearer about what to do when workflow validation fails (it shows validate_workflow but doesn't specify error handling steps as explicitly as the node validation loop does). | 4 / 5 |
Progressive Disclosure | The skill references SEARCH_GUIDE.md, VALIDATION_GUIDE.md, and WORKFLOW_GUIDE.md appropriately, but no bundle files were provided to verify these exist. The main issue is that the SKILL.md itself contains too much detail that should be in those referenced guides—the Unified Tool Reference section, detailed Template Usage, and extensive Common Mistakes sections duplicate what should be in the sub-guides, making the overview too long while still referencing external files. | 3 / 5 |
Total | 13 / 20 Passed |