Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    n8n MCP Workflow Architect

    1

    Build, repair, and safely activate n8n workflows through the n8n-mcp server with a discover-validate-patch loop that eliminates config and identifier errors.

    Free

    5 usesSecurity scanned
    n8n MCP Workflow Architect

    n8n MCP Workflow Architect

    Example session with this skill installed

    My n8n-mcp server is connected to our self-hosted instance. I have a saved workflow called Invoice Sync (id 17) that broke after an upgrade — the Slack node keeps erroring and the IF branches lead nowhere. Please use n8n-mcp to look up the current Slack node version and its schema, validate each node config against a strict profile, then patch the workflow with partial-update operations including proper branch connections, and only activate it once validate_workflow comes back clean. Include the workflowNodeType form of every node you touch so I can reuse the payload on a second instance, and note which fixes came from auto-sanitization versus your explicit operations.

    • Read your context and instructions
    • Compiled the n8n mcp workflow

    Summary

    Repaired the 'Lead Intake to Slack' workflow (id 42) via n8n-mcp: converted failing node types to the workflow identifier family, bound the orphaned IF branches with semantic parameters, and revalidated before activation.

    MetricBeforeAfter
    validate_workflow errors70
    Broken IF connections2 unbound2 (branch: true / branch: false)
    Identifier failures4 (nodes-base.* in create payload)0 (n8n-nodes-base.*)
    Activation statusFailed pre-flightActive, post-validation green

    Fixes were applied through n8n_update_partial_workflow with intent 'Bind IF branches after Switch refactor'. Auto-sanitization normalized two filter payloads on the Edit Fields nodes; the remaining corrections were explicit patch operations, each revalidated with validate_node profile runtime. Slack and HTTP nodes now use workflowNodeType n8n-nodes-base.slack / n8n-nodes-base.httpRequest for reuse on the second instance.

    Next steps

    • Attach Slack credentials and fire one live test at the /incoming webhook path
    • Add an error-output branch from the HTTP Request node before high-volume traffic
    • Re-run n8n_validate_workflow on a schedule to catch drift after n8n upgrades

    Connects securely to your tools. The creator never sees your data.

    About this skill

    The problem

    Building n8n workflows through LLMs often results in broken JSON, incorrect node type identifiers, and invalid connection logic. Manually debugging these configuration errors in the n8n UI is slow and disrupts the development loop.

    What it does

    • Discovers available n8n nodes using keyword and capability searches.
    • Validates node configurations against standard, runtime, or AI-friendly schemas before deployment.
    • Performs partial workflow updates to add, rename, or connect nodes without rewriting the entire graph.
    • Deploys tested templates from the n8n library with automatic version upgrading.
    • Runs health diagnostics and validation checks on existing workflows via the n8n API.

    Frameworks & tools

    n8n, n8n-mcp server, REST API, Node-RED style logic, and JSON-based workflow schemas.

    Why this beats prompting it yourself

    Basic LLMs frequently hallucinate node parameters or use the wrong naming conventions for different n8n versions. This skill implements a strict discover-validate-patch loop that verifies every change against the actual n8n-mcp server, ensuring the generated workflow is structurally sound and ready for activation.

    Use cases

    • Architecting complex branching logic using IF and Switch nodes with semantic connections.
    • Migrating or upgrading legacy n8n workflows by comparing node versions and breaking changes.
    • Automating the setup of standard integrations like Slack webhooks or database listeners.
    • Debugging existing production workflows by inspecting state and applying targeted patches.

    Known limitations

    Requires a configured n8n-mcp server. API-dependent tools require a valid n8n API URL and Key. Does not support fixing corrupted JSON blobs that fall outside the n8n schema.

    Tool parameter names can drift between n8n-mcp releases, so a step may need the tools_documentation lookup. Validation checks structure and configs, not runtime behavior — it cannot test-run a workflow inside n8n.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    5 uses

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 2 months ago

    • Free with an Agensi account

    Requires

    N8n

    Permissions

    Network Access

    Allowed Hosts

    n8n.cloud

    Network access is limited to your own n8n instance (self-hosted or n8n.cloud) reached through the already-configured n8n-mcp server.

    Listed2 months ago

    What's inside

    Frequently Asked Questions

    This skill provides a structured "discover-validate-patch" loop that allows your AI agent to browse actual node libraries, verify JSON schemas, and deploy workflows through the n8n-mcp server without manual copy-pasting or syntax errors.

    You will need a running n8n instance (self-hosted or Cloud), a valid n8n API key, and the n8n-mcp server configured to allow communication between your AI agent and the n8n environment.

    Yes, the skill includes tools for health diagnostics and partial updates, allowing your AI to inspect existing workflows, identify broken connections or outdated nodes, and apply targeted patches to fix them.

    It is compatible with any AI agent or platform that supports the Model Context Protocol (MCP), such as Claude Desktop or specialized IDE agents, provided they are connected to the n8n-mcp server.

    The Architect handles versioning by discovering the specific node requirements of your n8n instance and automatically upgrading or mapping node parameters to match the current version schema.

    New in Workflow Automation