Eight point diagnostic check covering Node, SDK, env vars, keypair, RPC latency, fallback RPC, SAP program, and disk. Run before deploys and after upgrades.

`doctor`

When something goes wrong, run doctor first.

`doctor run`

synapse-sap doctor run
synapse-sap doctor run --quick     # skip latency probes
synapse-sap doctor run --save out/doctor.json

The check covers eight areas:

Check	What it verifies
Node version	Matches the supported range (>= 18.17)
SDK version	Compatible with the CLI (`@oobe-protocol-labs/synapse-sap-sdk` peer range)
Environment	All required `.env` variables present and non empty
Keypair	File exists, has 0600 permissions, parses to a valid 64 byte secret
RPC reachability	Primary endpoint responds within timeout, with measured latency
Fallback RPC	Secondary endpoint responds and returns the same slot
SAP program	Program account exists at `SAPpUhsWLJG1FfkGRcXagEDMrMsWGjbky7AyhGpFETZ`
Disk space	At least 100 MB free for artifacts, cache, and logs

Exit codes

Code	Meaning
0	All checks passed
1	One or more warnings
2	One or more failures (CI should treat as fatal)

Output shapes

Human readable table by default. Machine output via --json:

synapse-sap doctor run --json | jq '.checks[] | select(.status != "ok")'

Each check is { id, label, status, detail, durationMs }. Use this in CI to fail fast and surface the precise failure reason in the build log.

When to run

Before the first command on a new machine.
After upgrading the CLI or the SDK.
As part of a CI pipeline before any state mutating step.
During incident response, before assuming the bug is in your code.

doctor