n8n is a powerful open-source workflow automation engine. K0nsult extends n8n with an enforceable governance layer — policy gates, audit trails, compliance validation, and human escalation built directly into the workflow execution pipeline.
This integration does not replace n8n. It wraps it with accountability.
"n8n users get workflow automation. K0nsult users get governed workflow automation — with audit trail, compliance, and human oversight built in. No need to rebuild workflows. Just add K0nsult as the control layer."
Data flows bidirectionally between n8n and K0nsult at five integration points:
n8n Workflow Engine K0nsult Governance Layer
┌──────────────────────┐ ┌───────────────────────────┐
│ │ │ │
│ Trigger │────────>│ Policy Gate │
│ │ │ (approve / deny / review)│
│ Nodes │ │ │
│ │<────────│ Decision Engine │
│ Execution │ │ (rules + risk scoring) │
│ │────────>│ │
│ Output │ │ Audit Trail │
│ │<────────│ (immutable log) │
│ Error Handling │ │ │
│ │────────>│ Escalation Router │
│ │ │ (human review queue) │
└──────────────────────┘ └───────────────────────────┘
│ │
│ Webhook Callbacks │
│<──────────────────────────────────>│
│ REST API (JSON) │
└────────────────────────────────────┘
Before n8n runs a workflow, K0nsult checks: Is this workflow approved? Does the user have permission? Is the data classification compatible? If any check fails, execution is blocked and the reason is logged.
K0nsult monitors n8n execution in real-time via webhook callbacks, flagging anomalies such as unexpected data volumes, unauthorized external calls, or execution time exceeding thresholds.
Every n8n execution is logged in K0nsult's immutable audit trail with the full decision chain: who triggered it, what policies were evaluated, what data was processed, and what output was produced.
When n8n hits uncertainty (ambiguous input, policy conflict, high-risk classification), K0nsult routes the decision to a human review queue. Execution pauses until the human approves, modifies, or rejects.
Before output is delivered, K0nsult validates it against policy rules and regulatory requirements. Non-compliant outputs are blocked, flagged, and logged with the specific rule violation.
K0nsult exposes five REST endpoints specifically for n8n integration:
POST /api/n8n/pre-check — validates workflow before execution
POST /api/n8n/execution-log — logs execution result to audit trail
POST /api/n8n/escalate — routes decision to human review queue
GET /api/n8n/policy/:id — returns policy rules for a specific workflow
POST /api/n8n/audit — records immutable audit trail entry
| Endpoint | Method | Request Body | Response |
|---|---|---|---|
/api/n8n/pre-check |
POST | workflow_id, user_id, input_data_class, trigger_type | { approved: bool, reason: string, policy_id: string } |
/api/n8n/execution-log |
POST | workflow_id, execution_id, status, duration_ms, output_summary | { logged: bool, audit_id: string } |
/api/n8n/escalate |
POST | workflow_id, execution_id, reason, context, urgency_level | { escalation_id: string, assigned_to: string, eta_minutes: int } |
/api/n8n/policy/:id |
GET | n/a (path param) | { policy: { rules: [], thresholds: {}, required_approvals: int } } |
/api/n8n/audit |
POST | action_type, actor_id, resource_id, decision, evidence | { audit_id: string, timestamp: ISO8601 } |
Configure these webhook nodes in your n8n workflows to connect to K0nsult:
| n8n Node Type | When | K0nsult Endpoint | Behavior |
|---|---|---|---|
| HTTP Request | Workflow start (before first action node) | /api/n8n/pre-check |
If response.approved === false, halt workflow and return reason |
| HTTP Request | On workflow complete | /api/n8n/execution-log |
Log execution result; continue regardless of response |
| HTTP Request | On error (error trigger node) | /api/n8n/escalate |
Escalate to human review queue; optionally wait for resolution |
| HTTP Request | Before output delivery | /api/n8n/audit |
Record final output in audit trail with compliance check |
| HTTP Request | Workflow initialization | /api/n8n/policy/:id |
Load policy rules into workflow context for conditional logic |
Tip: Use n8n's "IF" node after the pre-check call to branch on the approved field. This keeps your workflow logic clean and separates governance from business logic.
n8n Trigger fires
│
▼
HTTP Request ──> K0nsult /api/n8n/pre-check
│
├── approved: true ──> Continue workflow
│ │
│ Execute nodes
│ │
│ ┌────┴────┐
│ Success Error
│ │ │
│ ▼ ▼
│ /execution /escalate
│ -log (human review)
│ │ │
│ ▼ ▼
│ /audit Wait for
│ │ resolution
│ ▼ │
│ Deliver ┌───┘
│ output │
│ ▼
│ Retry or
│ terminate
│
└── approved: false ──> Block + Log reason
│
▼
/audit (denied)
n8n processes incoming invoices via email trigger, extracts data with AI, and routes for payment. K0nsult enforces approval thresholds: invoices under 5,000 EUR are auto-approved (Level A), 5,000-50,000 EUR require manager approval (Level C), and above 50,000 EUR require dual authorization (Level D). Every decision is logged with full audit trail for regulatory compliance.
n8n orchestrates multiple AI models for content classification: sentiment analysis, toxicity detection, and brand safety scoring. K0nsult enforces content policy gates — flagged content is routed to human moderators via the escalation bridge. No content reaches production without passing K0nsult's compliance gate. Full audit trail proves moderation diligence.
n8n handles ticket routing: classifying issues, assigning agents, and triggering automated responses. K0nsult ensures SLA compliance by monitoring response times and escalating breaches. Priority tickets are automatically routed through the escalation bridge. The audit trail provides evidence for SLA reporting and customer satisfaction metrics.
| Use Case | n8n Handles | K0nsult Enforces | Outcome |
|---|---|---|---|
| Finance | Invoice extraction, routing | Approval thresholds, audit trail | Compliant payment processing |
| Content | AI model orchestration | Policy gates, human review | Safe content delivery |
| Support | Ticket classification, routing | SLA compliance, escalation | Provable SLA adherence |
| Layer | Mechanism | Details |
|---|---|---|
| Transport | TLS 1.3 | All API calls encrypted in transit; certificate pinning recommended |
| Authentication | API Key + HMAC | Each n8n instance gets a unique API key; requests signed with HMAC-SHA256 |
| Authorization | Scoped permissions | API keys scoped to specific workflows and endpoints; least-privilege model |
| Rate limiting | Per-key throttle | 1000 req/15min per IP; 5 req/15min for forms; 429 response on exceed |
| Audit | Immutable logging | All API calls logged; INSERT-only PostgreSQL; no DELETE/UPDATE on audit tables |
| Without K0nsult | With K0nsult |
|---|---|
| Workflows run without oversight | Every execution is policy-checked and logged |
| No audit trail for decisions | Immutable, timestamped decision records |
| Errors go to logs | Errors escalate to humans with context |
| Compliance is self-reported | Compliance is system-generated evidence |
| Policy changes require code edits | Policy changes are instant rule updates |
| No kill switch for runaway automations | Single API call suspends all execution |
"n8n can build any workflow. But it cannot enforce who approves what, log why decisions were made, or prove compliance to a regulator. K0nsult can."