Home

Workload 0xeb2365914c7f6be3399c8d1a3a04f7d2ae87ad671089b0fdd0363b8ce1651fb5

Overall status: verified.

Download provenance bundle (.tar)

Source info

TDX measurements

MRTD0xc1ee9c16e3afc506cfe042c5b846a368528f3b37618eafb27469bc114cf914e9222c91618470e7f2b28ac360968270a5
RTMR00x7ccc8383a31967901f199d4e5bf0ead438492e6ce5f586253580ef024f08af0698e66492c5008856a7bed7078ea9b1ef
RTMR10x2d1f0da6067d854a185f8e8fecbc0c4eb5635247c6fa06f0b011ab72464021c9efe33020b33c6d008bf7d1912fa2c4f0
RTMR20x000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
RTMR30xe5e90eeba2131a8d0e39c3839660870eb7f56c8dedd2b7e31fa30aaee7f68df92c7cc43b3ff141a14029cac46e8de4d2
MRCONFIGID0x000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000
XFAM0xe702060000000000
TD attributes0x0000001000000000

Measured workload inputs

These exact files are inputs to the TDX measurement calculation for this workload.

Docker container images and source

Base VM source

ritual-net/ritual-images@422ff3db4158201040c63f20f2743447c29e0a73 · OS build instructions referenced by the platform profile

Platform profile

This workload records the opaque ID of the authority-accepted platform profile required to verify it.

pp_7623ccd8-536f-4464-a284-ca99e060fae7
Download platform-profile bundle (.tar)

Detailed verification checks

Follow the links from the workload ID to its measurements, images, and source code. Expand a check to see what it establishes and inspect the supporting evidence.

Provenance record

The provenance manifest is the central index for this workload. It records the expected TDX measurements, the locator and digest of each verification artifact, and the repository and commit for relevant source code. A verifier uses these references to obtain each available artifact, confirm that its bytes match the recorded digest, and supply the verified artifact to the corresponding verification check.

Download provenance manifest (JSON)

1. Start with an on-chain workload ID

You select the workload ID. This report does not check its current on-chain capability assignment.

Workload ID being checked: 0xeb2365914c7f6be3399c8d1a3a04f7d2ae87ad671089b0fdd0363b8ce1651fb5

2. Look up the TDX measurements

Check the provenance record and confirm that its eight measurement values produce the selected workload ID.

3. Reproduce the OS + Docker → TDX measurements

Use the verified prebuilt OS image, Docker images, Docker configuration, and TDX boot profile to recompute the eight TDX values. Compare each value with the expected measurements from step 2.

Verify the prebuilt OS image

This check links the published OS artifact to its recorded source repository, commit, and build instructions. The service hashes the build instructions and OS artifact, compares them with the provenance manifest, and confirms that the recorded source and artifact values match an immutable platform-profile bundle selected from the service-owned accepted-profile registry. The workload record names the selected profile’s opaque ID; the registry binds that ID to the exact archive. Each check below links the exact files and source used, so you can inspect them or repeat the comparisons.

For GCP/mkosi, the provider profile is accepted in a separate offline phase, where the service rebuilds the recorded ritual-images source commit and confirms that the rebuilt guest disk matches the published image.

Verify the Docker workload configuration used in the TDX measurements

This check links the Docker workload inputs to the TDX measurements that commit to them. The service checks the exact measured configuration against its manifest digest, extracts its digest-pinned container images, and confirms that the image list exactly matches the Docker images declared in the manifest. This means the GitHub attestations in step 4 cover the container images committed to the TDX measurements. The provider-specific details below explain which configuration is measured and how it contributes to the TDX values.

For GCP/mkosi, the measured configuration is the exact labeled byte stream containing the environment file followed by the Docker Compose file. Here, we are validating the connection between docker images -> full configuration, which can then be connected to the RTMR3 value. The service publishes the measured byte stream and both component files so you can inspect the exact inputs or repeat the calculation.

Verify the TDX boot profile

The TDX boot profile is the provider-specific recipe that combines the selected VM image with the cloud platform’s firmware, virtual-machine configuration, VMM behavior, and TDX launch settings to reproduce the boot-related measurements. This check links the boot-related TDX measurements back to the boot profile that produced them. The service verifies the supplied artifacts against the provenance manifest and requires the provider-specific profile values to match an accepted provider profile.

For GCP/mkosi, the accepted boot profile connects MRTD to the exact Google OVMF firmware and machine topology. It also supplies the calibration baseline used with the OS image’s UKI and GPT disk layout to calculate an expected RTMR1. The workload CCEL separately records the boot events observed for RTMR0, RTMR1, and RTMR2. The complete-tuple procedure below performs both calculations and requires the two RTMR1 values to match. A different cloud provider, machine type, or firmware version requires a different accepted boot profile.

Google OVMF firmware + machine topology        → MRTD inputs
workload CCEL                                   → observed RTMR0, RTMR1, RTMR2
OS image UKI + GPT layout + accepted baseline   → expected RTMR1
offline-approved VMM launch configuration       → MRCONFIGID, XFAM, TDATTRIBUTES

Reproduce TDX measurements

With the OS image, Docker configuration, and TDX boot profile established above, this section runs the provider-specific measurement procedure that ties them together. We use pinned binaries that calculate the resulting TDX measurements without needing to run a TDX machine, and link to their source code.

verified OS image + Docker configuration + TDX boot profile
                              ↓
                   pinned measurement tool
                              ↓
              recomputed eight-field TDX tuple
                              ↓
          canonical output digest matches manifest

For GCP/mkosi, the pinned cvm-measure implementation and Ritual adapter calculate MRTD from the endorsed firmware and machine topology, replay RTMR0, RTMR1, and RTMR2 from the workload CCEL, independently cross-check RTMR1 against the exact UKI and GPT disk layout, and extend RTMR3 with the SHA-384 digest of the Docker configuration.

All eight values match?

4. Verify each GitHub attestation

Does each exact container image connect to the claimed source code and build workflow?

5. (Optional) Rebuild the OS image

An independent OS rebuild is an offline step for the user. Matching the recorded OS to an approved profile in step 3 does not mean this request rebuilt it.

6. (Optional) Source audit and capability review

Reviewing whether the source is safe and appropriate for its capability is a human judgment. This report does not perform that audit.

All verification files

This appendix lists every verification input. Retained files and directly addressable external references are linked.

InputLocator
OS build instructionsevidence/base-build-definition
OS imagehttps://storage.googleapis.com/prod-mainnet-01-provenance-base-images/ritual-container-pr12-422ff3db4158-44c022d8.tar.gz
Docker configurationcdi-mkosi-measured-configuration.bin
Measurement tool build instructionshttps://raw.githubusercontent.com/cohere-ai/cvm-measure/7c2e57fd546f5f25056376e936adccf498e9ef81/pyproject.toml
Measured environmentcdi-mkosi-environment
Measured Docker Compose filecdi-mkosi-compose.yml
Platform firmwarehttps://storage.googleapis.com/gce_tcb_integrity/ovmf_x64_csm/53c558fe06714203d7428484c46ee7c0b9df748a9789e173421c28d097115b30efb8476304617532488ab40c3122db88.fd
Calibration baselineevidence/baseline
Calibration boot event logevidence/calibration-ccel
Workload boot event logcdi-mkosi-ccel.bin
GCP TDX machine profilecdi-mkosi-machine-profile.json
claw-spawner — Docker image metadatadocker.io/ritualnetwork/claw-spawner:563c266-amd64@sha256:836eae3b994b60b0c08bedd212dfa64d04625e12f87ccf9c1292ccde8316d261
docker-socket-proxy — Docker image metadatadocker.io/ritualnetwork/docker-socket-proxy:v0.4.2-2f04313-563c266-amd64@sha256:5c6ece5b10393237ad5ce264a06496d8bc41345be19d61eb6744dbce2cac7994
executor-go — Docker image metadatadocker.io/ritualnetwork/executor-go:7929fd8-0e66c99-amd64@sha256:2ff80a98f0c1e26a0da1f0cc5f2c2d238bc0d47fdde7f028ab030ead68165734
hermes-instance — Docker image metadatadocker.io/ritualnetwork/hermes-instance:f90a8c9-563c266-amd64@sha256:eda1412b8cb693daa17b8b9636a47a0f002cac782e87cbd47e8416cce07bee91
openclaw-instance — Docker image metadatadocker.io/ritualnetwork/openclaw-instance:563c266-amd64@sha256:06237d4b67195d095959b08bcf06d7ccc72f5b29e82406f716bc7374e01ae856
ritual-pii — Docker image metadatadocker.io/ritualnetwork/ritual-pii:7929fd8-0e66c99-amd64@sha256:7779a257fc54873c2d77f4373163a877f4085ef235c937283647148b64c0edaa
sovereign-agent — Docker image metadatadocker.io/ritualnetwork/sovereign-agent:563c266-amd64@sha256:a7e7c0186c8c3569935eba1d94e9b4d50f90e3471333c1ebcacffd050f7f89f1
zeroclaw-instance — Docker image metadatadocker.io/ritualnetwork/zeroclaw-instance:502b5f2-563c266-amd64@sha256:1d2b2298e72f3503c060ca933f804f0c2d0c92d2a87feb99e90d7f7e14b043d2