Add KMS signer support to run-enclave.sh#433
Conversation
|
Claude finished @jjeangal's task in 1m 8s —— View job Review of PR #433: Add KMS signer support to run-enclave.sh
SummaryClean, well-scoped change. The three-way auth branch (KMS signer → private key → test mnemonic) is well-structured and fully backward-compatible. The One suggestion posted as an inline comment: If only one of Everything else looks good — the logging additions are clear, the flag names ( |
There was a problem hiding this comment.
Code Review
This pull request introduces support for KMS signing in the run-enclave.sh script by adding SIGNER_ENDPOINT and SIGNER_ADDRESS configuration options. The review feedback suggests simplifying the log messages for missing KMS variables to avoid confusion regarding fallback behavior. Furthermore, it is recommended to implement explicit error handling for partial KMS configurations to prevent accidental silent fallbacks to private keys or mnemonics.
…g log Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This PR
run-enclave.sh: whenSIGNER_ENDPOINTandSIGNER_ADDRESSenv vars are set, pass--signer.endpointand--signer.addressto the enclave batcher instead of--private-key.OP_BATCHER_PRIVATE_KEYor test mnemonic paths are unchanged.Key Places to Review
espresso/docker/op-batcher-tee/run-enclave.sh— authentication section (lines ~96-107): KMS signer branch added above the existing private key branch.How to Test This PR
enable_kms_signer = trueintee-op-deploy— the ECS task will setSIGNER_ENDPOINTandSIGNER_ADDRESSenv vars on the container.--signer.endpointand--signer.addressin its args (visible in logs).enable_kms_signer = falseare unaffected — the private key path runs as before.