Description
n8n-nodes-identark
This is an n8n community node. It lets you use IdentArk in your n8n workflows.
IdentArk is credential-isolation and governance infrastructure for AI agents. Your n8n
instance routes LLM calls through the IdentArk gateway, so the API keys live in IdentArk’s
vault — not in your workflows, environment, or credentials store. Every call is
risk-scored, cost-tracked, and written to an immutable audit log, and high-risk actions can
be paused for human approval.
n8n is a fair-code licensed workflow automation platform.
Installation
Operations
Credentials
Compatibility
Usage
Resources
Version history
Installation
Follow the installation guide in the n8n community nodes documentation.
In short: Settings → Community Nodes → Install, then enter n8n-nodes-identark.
The node is free and open source. It connects to the IdentArk control plane, which is where credential isolation, risk scoring, approvals, and audit logging run — create a free account at identark.io to get started.
Operations
| Operation | Description |
|———–|————-|
| Invoke LLM | Send a message to an LLM through the gateway and get the response. Supports system / user / assistant roles and continuing a conversation via a session ID. |
| Create Session | Open a credential-isolated agent session with a model, provider, vault credential reference, and a cost cap. |
| Get Session Cost | Read the running spend for a session. |
| Register Agent | Register a new agent identity with a default model, provider, and credential reference. |
| Delete Agent | Remove an agent identity. |
| List Agents | List registered agents, optionally filtered by active/inactive. |
| List Pending Approvals | List Human-in-the-Loop approvals awaiting a decision. |
| Get Approval | Fetch the full context of a single approval request. |
| Submit Decision | Approve or reject a pending request, with an optional comment and MFA token. |
Credentials
You need an IdentArk account and an API key. Create a free account at identark.io → (no card required).
1. Sign up at identark.io and create an organisation.
2. In the dashboard, open Settings → API Keys and generate a key (it starts with csk_).
3. Register your LLM provider credentials in the IdentArk vault.
Then in n8n create an IdentArk API credential:
| Field | Description |
|——-|————-|
| API Key | Your IdentArk API key (csk_…). Sent as a bearer token. |
| Control Plane URL | Base URL of your IdentArk control plane. Defaults to https://api.identark.io. |
When you save the credential, n8n calls GET /v1/agents against your control plane. This
validates the API key itself rather than only checking that the URL is reachable. The key
needs the agents:read scope, which is included in the read, invoke, and admin
presets.
Compatibility
- Requires n8n with community-node API version 1.
- Tested against the current n8n LTS release.
- Node.js >= 20.15.
- No runtime dependencies.
- n8n community nodes documentation
- IdentArk documentation
- IdentArk SDK on GitHub
Usage
A typical governed-agent workflow:
1. Create Session — choose the model, provider, vault credential reference, and cost cap.
2. Invoke LLM — pass the returned session ID to continue the conversation. The provider
key never enters n8n; the gateway injects it at call time.
3. Get Session Cost — check spend before the next step or for billing.
To add human oversight, branch on List Pending Approvals → Get Approval →
Submit Decision: surface high-risk actions to a reviewer (e.g. via a Slack or email
node) and only proceed once the decision is approved.
If you’re new to n8n, see Try it out.
Resources
Version history
1.0.0
Initial release: Invoke LLM, Create/Get-cost Session, agent management
(Register/Delete/List), and the Human-in-the-Loop approval flow
(List Pending / Get / Submit Decision).