Certify any AI API call with cryptographic proof of execution — directly from your n8n workflows.
The only trust layer that works across any model, any provider, any infrastructure. In a real-world pipeline (Claude + Mistral + GPT + custom APIs), no single vendor can certify the full chain. ArkForge Trust Layer can.
- Go to Settings > Community Nodes
- Select Install a community node
- Enter
n8n-nodes-arkforge - Agree to the risks and click Install
cd ~/.n8n
npm install n8n-nodes-arkforge- Get your free API key at arkforge.tech
- In n8n, go to Credentials > New > ArkForge Trust Layer API
- Paste your API key
Proxy any API call through the Trust Layer. The upstream API is called normally, and you get back:
- The original API response (unchanged)
- A cryptographic proof with SHA-256 hashes of request, response, and chain
- A verification URL anyone can visit to verify the proof
- An optional RFC 3161 timestamp from an independent authority
Fields:
| Field | Required | Description |
|---|---|---|
| Target URL | Yes | The upstream API to call (e.g. https://api.anthropic.com/v1/messages) |
| HTTP Method | No | POST (default) or GET |
| Payload | Yes | JSON body to send to the upstream API |
| Extra Headers | No | Headers forwarded to the upstream API (e.g. Authorization: Bearer sk-...) |
| Description | No | Metadata attached to the proof |
| Agent Identity | No | Agent identity recorded in the proof |
| Agent Version | No | Agent version recorded in the proof |
Look up and verify an existing proof by its ID.
Fields:
| Field | Required | Description |
|---|---|---|
| Proof ID | Yes | The proof ID (e.g. prf_20260312_155129_11b6cb) |
The Certify operation returns:
{
"proof": {
"proof_id": "prf_...",
"verification_url": "https://arkforge.tech/trust/v1/proof/prf_...",
"hashes": {
"request": "sha256:...",
"response": "sha256:...",
"chain": "sha256:..."
},
"timestamp": "2026-03-12T15:51:29Z",
"transaction_success": true
},
"service_response": {
"status_code": 200,
"body": { /* original API response */ }
}
}Access the upstream response: {{ $json.service_response.body }}
Access the proof ID: {{ $json.proof.proof_id }}
Access the verification URL: {{ $json.proof.verification_url }}
- Multi-LLM pipelines: Certify every step when Claude calls Mistral calls GPT
- EU AI Act compliance: Prove which models processed what data, and when
- Audit trails: Immutable proof of every AI decision in regulated workflows
- Supply chain trust: Verify that API responses haven't been tampered with