SecretVM TEE quickstart

This guide walks you through deploying a Morpheus TEE-hardened provider node on SecretVM (SCRT Labs’ confidential VM platform). By the end you’ll have a provider running inside a hardware-secured Intel TDX enclave that consumers can cryptographically verify before sending prompts. Most of the operator work after deploy is graphical: MyProvider against your SecretVM HTTPS URL, plus looking up models on active.mor.org/status. Swagger remains a fallback. For deeper details (cosign verification, RTMR3 recomputation, attestation manifest inspection) see TEE reference. For the trust-chain conceptual picture see TEE overview.

What you’ll need

A funded wallet with MOR + ETH on Base Mainnet (or Base Sepolia for testnet).
The wallet’s private key — this stays encrypted inside the TEE.
An RPC endpoint for Base Mainnet (e.g. wss://base-mainnet.g.alchemy.com/v2/<your_key>).
Your AI model backend reachable via a private URL (e.g. http://my-model:8080/v1/chat/completions).
A SecretVM account at https://secretai.scrtlabs.com — sign up to get an API key.
(Recommended) secretvm-cli: sudo npm install --global secretvm-cli.
(Recommended) MyProvider for registration and bidding after the VM is healthy.

Step 0: Pick the model you will serve

Before minting anything on chain, see what already exists and at what price:

curl -sS https://active.mor.org/active_models.json \
  | jq '.models[] | select(.Name | test("your-model-name"; "i")) | {Name, Id, Tags, bidDetail}'

curl -sS https://active.mor.org/active_bids.json \
  | jq '[.bids[] | select(.ModelName | test("your-model-name"; "i"))] | sort_by(.PricePerSecond | tonumber) | .[:5]'

If a matching model Id already exists and (for TEE) it carries a tee tag — plan to bid on that Id and put it in MODELS_CONFIG_CONTENT below.
If you need TEE and no suitable tee-tagged model exists — you will register a new model with tag tee after deploy (Step 5).
Or browse the human UI: active.mor.org/status.

Full decision table: Register on chain → Step 0.

Step 1: Get the Docker Compose file

Each CI/CD build produces a deployed compose file pinning the TEE image by SHA-256 digest — that’s what SecretVM uses to compute RTMR3. Download docker-compose.tee.deployed.yml from the latest GitHub Release — the digest is already filled in. You can also use the template at proxy-router/docker-compose.tee.yml and replace the image tag yourself. The compose includes a Traefik TLS sidecar that uses SecretVM’s auto-generated certs to serve the API on HTTPS port 443. The full compose body is reproduced in TEE reference → Step 1.

Use the digest-pinned version. Tags like :latest are mutable; the digest is not. Using the digest ensures RTMR3 will match the CI/CD-computed golden value.

Step 2: Prepare your secrets

Create a .env file with your 5 provider-specific secrets:

WALLET_PRIVATE_KEY=<your_private_key>
ETH_NODE_ADDRESS=wss://base-mainnet.g.alchemy.com/v2/<your_alchemy_key>
MODELS_CONFIG_CONTENT={"models":[{"modelId":"0x<your_model_id>","modelName":"your-model-name","apiType":"openai","apiUrl":"http://your-model:8080/v1/chat/completions","concurrentSlots":6,"capacityPolicy":"simple"}]}
WEB_PUBLIC_URL=https://your-public-domain.com
COOKIE_CONTENT=admin:<your_secure_password>

MODELS_CONFIG_CONTENT must be single-line JSON. Schema: models-config.json. If Step 0 already gave you an existing on-chain Id, put that value in modelId now. If you will mint a new tee model after deploy, use a placeholder and replace it via MyProvider Model Configuration Sync (Step 5) before posting a bid you care about.

Step 3: Deploy on SecretVM

Web portal
CLI (recommended)

Go to https://secretai.scrtlabs.com/secret-vms/create
Docker Compose: paste the contents of your compose file.
Encrypted Secrets: enter your 5 variables.
Advanced features:

Setting	Recommended	Notes
Platform	Intel TDX	Required to match CI/CD golden values
Additional Files	Empty	Adding files changes RTMR3
Enable Persistence	On	Preserves data across reboots
Enable Upgrades	On	Allows updates without reprovisioning
Hide Runtime Info	Off	Keeps the `/docker-compose` endpoint accessible for verification

Click Deploy.

secretvm-cli auth login

secretvm-cli -k <your_api_key> vm create \
  --name morpheus-provider \
  --type small \
  --persistence \
  --upgradeability \
  --docker-compose proxy-router/docker-compose.tee.yml \
  --env path/to/your/.env \
  --platform tdx

secretvm-cli -k <your_api_key> vm list

Full CLI reference: SecretVM CLI docs.

Step 4: Verify the node is running

Once provisioned (typically 2-3 minutes):

curl https://<your-secretvm-url>/healthcheck

The API is served on HTTPS port 443 via Traefik. Expected:

{ "status": "healthy", "version": "v7.0.0", "uptime": "1m30s" }

Verify the attestation endpoint (separate port, direct from SecretVM):

curl -k https://<your-secretvm-url>:29343/cpu | head -c 100

Returns a long hex string — your hardware-signed TDX attestation quote.

Step 5: Register via MyProvider (recommended)

SecretVM already terminates TLS on port 443, so the hosted GUI at myprovider.mor.org can talk to your node without mixed-content issues.

Open MyProvider

Go to myprovider.mor.org. Connect with:

URL: https://<your-secretvm-url>/
Username / password: from COOKIE_CONTENT (e.g. admin:yourpassword)

Provider tab → verify the public :3333 endpoint → create provider (min 0.2 MOR stake). Approve Diamond spend if prompted.

Bid — prefer an existing model

If Step 0 found a suitable model (including tee when you need TEE): Models & Bids → Available Models → Add Bid using that Id. Set price from the competitor check.

Only if needed — create a tee-tagged model

If no suitable tee model exists, use Create Model & Bid and include tag tee (comma-separated tags field). The tee tag is what triggers consumer-side attestation; without it, consumers treat you as a standard provider.

Sync models-config

Use Model Configuration Sync to generate updated MODELS_CONFIG_CONTENT, then update SecretVM encrypted secrets and redeploy/restart if the on-chain modelId changed from your Step 2 placeholder. Details: MyProvider GUI.

Swagger fallback: https://<your-secretvm-url>/swagger/index.html — same contract calls as Register on chain. Click Authorize with your COOKIE_CONTENT / .cookie credentials first (API auth).

Step 6: Verify your attestation

Quick check (SecretVM portal)

Go to https://secretai.scrtlabs.com/attestation
Paste your compose file contents.
Enter your VM URL.
Click Verify.

Three layers must pass: hardware (genuine Intel TDX), VM (known SecretVM firmware/kernel), software/RTMR3 (your image + compose match what CI/CD built).

Programmatic check (cosign)

cosign verify \
  --certificate-oidc-issuer https://token.actions.githubusercontent.com \
  --certificate-identity-regexp 'MorpheusAIs/Morpheus-Lumerin-Node' \
  ghcr.io/morpheusais/morpheus-lumerin-node-tee:<your-version>

cosign verify-attestation \
  --certificate-oidc-issuer https://token.actions.githubusercontent.com \
  --certificate-identity-regexp 'MorpheusAIs/Morpheus-Lumerin-Node' \
  --type https://morpheusais.github.io/tee-attestation/v1 \
  ghcr.io/morpheusais/morpheus-lumerin-node-tee:<your-version> \
  2>/dev/null | jq -r '.payload' | base64 -d | jq '.predicate'

What consumers see, what your P-Node does

The two-hop trust chain — see TEE overview for the full conceptual breakdown.

C-Node (v6.0.0+) ─Phase 1─▶ P-Node (-tee, v7.0.0+) ─Phase 2─▶ Backend LLM (SecretVM)

Phase 1 (consumer ↔ your P-Node): the consumer’s proxy-router fetches your :29343/cpu quote, verifies via SecretAI portal, checks TLS binding, compares RTMR3 against the cosign-signed golden values, and re-checks with a ~50 ms fast path on every prompt.
Phase 2 (your P-Node ↔ backend): your P-Node resolves the backend’s attestation endpoint (probes :21434 first, falls back to :29343), fetches /cpu, verifies the TLS binding (SPKI or full-certificate digest in reportData), replays workload RTMR3 from /docker-compose, verifies GPU evidence via NVIDIA NRAS (mandatory — a failed or unreachable NRAS fails attestation), and re-checks per prompt. Per-model state is exposed at GET /v1/models/attestation.

Forward compatibility: a v6.0.0+ C-Node only needs to understand Phase 1; Phase 2 is your P-Node’s job. Any v6+ consumer works with any v7+ provider, gaining Phase 2 transparently.

Troubleshooting

Symptom	Likely cause	Fix
Healthcheck returns nothing	VM still provisioning	Wait 2-3 min, check `secretvm-cli vm list`
Attestation quote empty	Port 29343 not exposed	SecretVM exposes it automatically — check VM status
RTMR3 mismatch	Wrong compose content or rootfs version	Use the exact deployed compose from CI/CD artifacts (byte-for-byte)
TLS binding fails	Using a proxy/CDN that terminates TLS	The consumer must connect directly — no TLS-terminating intermediaries on `:29343`
`tee` model not getting sessions	Consumers on older versions	Consumers need v6.0.0+ for Phase 1
Phase 2 failing silently on consumer	Phase 2 runs in your P-Node	Check `GET /v1/models/attestation` on your node
Phase 2 fails with `NRAS returned status 403`	NVIDIA’s attestation edge blocks the node’s IP/region	Run the P-Node from a network (or VPN egress) with access to `nras.attestation.nvidia.com` — NRAS verification is mandatory

Updating your node

secretvm-cli -k <your_api_key> vm edit <your_vm_uuid> \
  --docker-compose path/to/new/docker-compose.tee.yml

The VM reboots with the new image. Verify with /healthcheck and the attestation portal.

Resource	Link
MyProvider (operator GUI)	MyProvider GUI
Discover models / bid	Register on chain
Live marketplace	active.mor.org/status
Full TEE reference	tee-reference
Conceptual TEE overview	TEE overview
models-config.json	reference
Verify end-to-end	Verify setup
Standard Docker (non-TEE)	proxy-router-docker
SecretVM documentation	https://docs.scrt.network/secret-network-documentation/secretvm-confidential-virtual-machines
SecretVM CLI	https://docs.scrt.network/secret-network-documentation/secretvm-confidential-virtual-machines/secretvm-cli
TEE images on GHCR	https://github.com/orgs/MorpheusAIs/packages?repo_name=Morpheus-Lumerin-Node

Get started

Inference API (hosted)

Concepts

Consumers

Prosumers

Providers - Resale

Providers - Full P-Node

AI knowledge

Ecosystem

Reference

What you’ll need

Step 0: Pick the model you will serve

Step 1: Get the Docker Compose file

Step 2: Prepare your secrets

Step 3: Deploy on SecretVM

Step 4: Verify the node is running

Step 5: Register via MyProvider (recommended)

Step 6: Verify your attestation

Quick check (SecretVM portal)

Programmatic check (cosign)

What consumers see, what your P-Node does

Troubleshooting

Updating your node

Further reading

​What you’ll need

​Step 0: Pick the model you will serve

​Step 1: Get the Docker Compose file

​Step 2: Prepare your secrets

​Step 3: Deploy on SecretVM

​Step 4: Verify the node is running

​Step 5: Register via MyProvider (recommended)

​Step 6: Verify your attestation

​Quick check (SecretVM portal)

​Programmatic check (cosign)

​What consumers see, what your P-Node does

​Troubleshooting

​Updating your node

​Further reading

What you’ll need

Step 0: Pick the model you will serve

Step 1: Get the Docker Compose file

Step 2: Prepare your secrets

Step 3: Deploy on SecretVM

Step 4: Verify the node is running

Step 5: Register via MyProvider (recommended)

Step 6: Verify your attestation

Quick check (SecretVM portal)

Programmatic check (cosign)

What consumers see, what your P-Node does

Troubleshooting

Updating your node

Further reading