What does this skill do, and when should you use it?
This is one Agent Skill from the n8n-skills repository, focused on operation-aware n8n node configuration. It explains how required fields change by resource and operation, how displayOptions create conditional fields, and when to request standard, full, or property-search node details. It covers configuration patterns for HTTP, Slack, database, conditional-logic, SplitInBatches, and Google Sheets nodes, with a validation-driven workflow. It is suited to users building workflows through n8n-mcp, but it does not replace skills for workflow architecture, coding, or n8n deployment.
Guides the use of get_node for standard or full node details and search_properties for targeted fields; identifies required configuration from resource, operation, and field dependencies; recommends starting with a minimal configuration and iterating through validate_node errors; explains when to use patchNodeField in n8n_update_partial_workflow for surgical text edits; and documents configuration cautions involving credentials, node UUIDs, typeVersion, empty database write outputs, and silent failure modes in several node families.
- An automation developer configuring Slack, HTTP Request, or database nodes through n8n-mcp and needing operation-specific required fields.
- A user configuring POST, PUT, or PATCH requests and resolving the sendBody-to-body dependency in an HTTP Request node.
- A workflow builder changing a Slack operation and needing to replace fields that belonged to the previous operation.
- A user editing a small portion of Code-node code, HTML email content, or a JSON body without replacing the entire field.
- A user configuring database writes, Switch, Merge, Webhook, or Schedule Trigger nodes and checking for runtime issues that validation may not catch.
How do you install this skill?
- Verify version-sensitive details such as timezone placement and database parameter fields against the target n8n instance schema before submission.
- Do not treat the stated success rates, evaluation count, or end-to-end verification as proven; the supplied material contains no supporting test evidence.
- Before deployment or workflow updates, separately confirm credentials, external writes, response behavior, and rollback procedures.
- Reachability of the n8n-mcp/Claude Code dependency from mainland-China networks is not established by the supplied material.
- MCP Server
n8n-mcp MCP server
Install and configure the n8n-mcp MCP server and have access to Claude Code, Claude.ai, or the Claude API. The recommended Claude Code command is: /plugin install czlonkowski/n8n-skills. For manual installation, run: git clone https://github.com/czlonkowski/n8n-skills.git, then copy n8n-skills/skills/n8n-node-configuration/ into the Claude Code skills directory. The source does not provide separate installation commands for other clients.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/czlonkowski/n8n-skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/n8n-node-configuration" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- How do I configure the HTTP Request node?
In an environment where this skill is loaded and n8n-mcp is available, ask a relevant question such as “How do I configure the HTTP Request node?” Start with get_node using standard detail, use search_properties when a field is unclear, and request full detail only when necessary. Configure the node, call validate_node, address the reported requirement, and repeat until valid.
What are this skill's strengths and limitations?
- Clearly separates resource, operation, and conditional field requirements.
- Provides a progressive discovery path through standard detail, property search, and full detail.
- Uses concrete HTTP, Slack, database, and conditional-logic examples with a validate-and-iterate workflow.
- Calls out practical issues involving credential placeholders, UUIDs, typeVersion, empty outputs, and silent runtime failures.
- Requires the n8n-mcp MCP server, which the skill does not provide.
- Primarily supplies configuration guidance; it is not a standalone node configurator or complete workflow generator.
- Some detailed guidance is delegated to related skills or repository documents not included in the supplied SKILL.md.
- No independent test results or cross-platform test evidence are provided for this individual skill.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| n8n Node Configuration Guide this page | 46 · Use with care | ★ 6.4k | 19d ago | MIT |
| n8n MCP Workflow Router | 50 · Use with care | ★ 6.4k | 19d ago | MIT |
| n8n Subworkflow Architect | 49 · Use with care | ★ 6.4k | 19d ago | MIT |
| n8n MCP Tools Expert | 45 · Not recommended | ★ 6.4k | 19d ago | MIT |
| n8n Error Handling | 53 · Use with care | ★ 6.4k | 19d ago | MIT |
How did FollowSkills review this skill?
The documentation shows meaningful security awareness through validation-first guidance, avoidance of placeholder credentials, parameterized SQL, silent-data-loss warnings, and explicit response-code handling. Points are deducted because deployment and workflow updates remain routine next steps without user confirmation, sensitive-data boundaries, rollback, change isolation, or complete disclosure of external effects; publisher identity and source attribution are also not independently verified.
The skill provides operation discovery, dependency inspection, iterative validation, and some diagnosable failure handling, including strict patchNodeField behavior. Points are deducted for visible inconsistencies: SKILL.md says timezone is workflow-level while OPERATION_PATTERNS.md places timezone inside rules; database parameter fields vary between queryReplacement and additionalFields.queryParameters; and several claims depend on unspecified tools, versions, or runtime behavior. No reproducible execution evidence is supplied, so the static cap keeps this at 10 or below.
README provides trigger phrases, common scenarios, tool context, and covered node categories, giving the skill a reasonably clear target for n8n configuration work. Points are deducted because non-fit boundaries, version-compatibility limits, and tool-availability requirements are incomplete; Chinese-language support is not declared, and mainland-China reachability of the n8n-mcp/Claude Code dependency is not established.
The material has a useful information hierarchy, progressive disclosure, related references, examples, gotcha sections, version history, MIT licensing, and author information. Points are deducted because maintenance ownership, issue/feedback channels, update procedures, and change tracking are weak; the README claims four evaluations without providing them here, while usage statistics and the end-to-end verification claim lack attached evidence.
The central workflow is clear and practically useful: inspect resource and operation, understand displayOptions, validate, and iterate. Examples cover common HTTP, Slack, database, and conditional-node cases. Points are deducted because some examples may be version-sensitive or contradictory, configurations still require instance-specific schema correction, and no statically verifiable representative output demonstrates deploy-ready completeness.
The files contain concrete configurations, field names, failure symptoms, and proposed fixes, which provide some auditability. Points are deducted because success rates, timing metrics, coverage claims, and end-to-end verification are author statements without committed tests, CI evidence, or independent corroboration; static review cannot reproduce them, so the score remains below the static ceiling of 5.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →