Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Answers n8n product, setup, credential, node, hosting, API, and usage questions from current n8n docs. Load n8n-docs via load_tool before calling it (search "n8n docs" if not visible). Use when the user asks how to configure, set up, troubleshoot, or understand n8n behavior, especially credential setup questions opened from the credential modal.
.claude/skills/n8n-io-n8n-docs-assistant/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | -52% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -59% | 0% |
| case-16 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-04 | ✓→✗ | ▼ Worse | -30% | 0% |
Use this skill when the user asks how to configure, set up, troubleshoot, or understand n8n behavior and the answer should come from current n8n docs.
n8n-docs(action="lookup") first for credential setup and direct n8ndocs questions. Use search then read only when you need tighter control over candidate pages.
intent: "credential-setup" and all availablecontext fields: credentialType, credentialDisplayName, documentationUrl, oauthRedirectUrl, and nodeType.
returned, prefer credential-specific pages over general credential UI pages.
Source: [Page title](page URL) when one docspage was used, or Sources: when multiple docs pages were used. Use only pages returned by n8n-docs.
OAuth Redirect URL from the modal unless oauthRedirectUrl is available in context.
credential modal, but never ask them to paste secrets into chat.
not navigate browser pages unless credential-setup-with-computer-use is explicitly needed and Computer Use browser tools are available.
If docs lookup fails, say that current docs could not be loaded. Provide only generic n8n UI guidance that is visible from the product context, and avoid claiming provider-specific setup steps without docs evidence.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 13,672 | 7,320 | -46% | 1 | 1 | 0% | 2,257 | 756 | -67% | 0 | 0 | — |
case-02 | fail→fail | 13,612 | 6,723 | -51% | 1 | 1 | 0% | 2,354 | 753 | -68% | 0 | 0 | — |
case-03 | fail→fail | 13,037 | 22,152 | +70% | 1 | 1 | 0% | 2,041 | 725 | -64% | 0 | 0 | — |
case-04 | pass→fail | 8,110 | 14,597 | +80% | 1 | 1 | 0% | 1,330 | 928 | -30% | 0 | 0 | — |
case-05 | pass→fail | 12,028 | 11,079 | -8% | 1 | 1 | 0% | 1,787 | 733 | -59% | 0 | 0 | — |
case-06 | fail→fail | 10,967 | 9,859 | -10% | 1 | 1 | 0% | 1,541 | 680 | -56% | 0 | 0 | — |
case-07 | fail→fail | 17,917 | 7,626 | -57% | 1 | 1 | 0% | 2,866 | 782 | -73% | 0 | 0 | — |
case-08 | fail→pass | 9,312 | 2,769 | -70% | 1 | 1 | 0% | 1,347 | 641 | -52% | 0 | 0 | — |
case-09 | pass→fail | 8,431 | 7,854 | -7% | 1 | 1 | 0% | 1,423 | 835 | -41% | 0 | 0 | — |
case-10 | fail→pass | 8,189 | 3,443 | -58% | 1 | 1 | 0% | 1,290 | 744 | -42% | 0 | 0 | — |
case-11 | pass→pass | 6,091 | 12,176 | +100% | 1 | 1 | 0% | 755 | 716 | -5% | 0 | 0 | — |
case-12 | pass→fail | 6,518 | 12,786 | +96% | 1 | 1 | 0% | 949 | 837 | -12% | 0 | 0 | — |
case-13 | pass→fail | 7,238 | 11,704 | +62% | 1 | 1 | 0% | 773 | 882 | +14% | 0 | 0 | — |
case-14 | pass→fail | 6,190 | 13,417 | +117% | 1 | 1 | 0% | 958 | 728 | -24% | 0 | 0 | — |
case-15 | fail→pass | 15,067 | 14,889 | -1% | 1 | 1 | 0% | 2,290 | 933 | -59% | 0 | 0 | — |
case-16 | fail→pass | 8,883 | 3,807 | -57% | 1 | 1 | 0% | 1,324 | 839 | -37% | 0 | 0 | — |
case-17 | fail→fail | 13,508 | 10,734 | -21% | 1 | 1 | 0% | 2,341 | 790 | -66% | 0 | 0 | — |
case-18 | pass→fail | 8,210 | 6,441 | -22% | 1 | 1 | 0% | 1,343 | 651 | -52% | 0 | 0 | — |
case-19 | fail→fail | 7,619 | 18,873 | +148% | 1 | 1 | 0% | 1,201 | 778 | -35% | 0 | 0 | — |
case-20 | fail→fail | 9,383 | 8,569 | -9% | 1 | 1 | 0% | 1,543 | 699 | -55% | 0 | 0 | — |
case-21 | fail→fail | 7,439 | 6,186 | -17% | 1 | 1 | 0% | 1,247 | 685 | -45% | 0 | 0 | — |
case-22 | pass→pass | 11,189 | 10,528 | -6% | 1 | 1 | 0% | 1,503 | 1,056 | -30% | 0 | 0 | — |
case-23 | pass→fail | 6,241 | 6,702 | +7% | 1 | 1 | 0% | 971 | 731 | -25% | 0 | 0 | — |
case-24 | pass→fail | 11,409 | 8,136 | -29% | 1 | 1 | 0% | 1,991 | 889 | -55% | 0 | 0 | — |
case-25 | pass→fail | 16,923 | 6,969 | -59% | 1 | 1 | 0% | 3,118 | 711 | -77% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 25 cases were attempted, and 6 counted toward the lift figure. The other 19 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of -24 percentage points is the difference between those two pass rates over the 6 comparable cases. 18 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.