@0xsteins is attempting to deploy a commit to the Ijai's projects Team on Vercel.

A member of the Team first needs to authorize it.

@0xsteins Great news! 🎉 Based on an automated assessment of this PR, the linked Wave issue(s) no longer count against your application limits.

You can now already apply to more issues while waiting for a review of this PR. Keep up the great work! 🚀

Learn more about application limits

Stong PR.... IMpressive!

Dispatch + upsert idempotency, deterministic IDs, structured skip-logging, contract-ID gating, and family-complete tests all hit the checklist. Two things before merge, then cleanup.

1. Tests skip the decode layer and positional field map — exactly where the risk is. so4Event injects a ready-made named object, so dispatch/idempotency are covered, but decodeTuple/decodeAddress/decodeBytesN32/decodeInteger and every fieldString(..., ) fallback are untested — and positional is the path that actually runs for tuple events. Please add decoder tests with real xdr.ScVal fixtures, one test running a raw event through decodeSorobanEvent with named empty, and verify the indices (sizeDeltaUsd@5, pnlUsd@11, …) against the real SO4 event signatures — right now they're assumptions.
2. namedTuple assumes Vec payloads are flat [key, value, …]. If SO4 emits positional tuples or ScMaps, that synthesizes junk named entries (masked today by the positional fallback). Confirm the real shape: ScMap is already handled; for Vec, drop the synthesis and go purely positional with documented indices.
   Cleanup:

SO4_CONTRACTS is hardcoded while a sync:contracts script exists — two sources of truth. If the literal is manual, import from the synced manifest so addresses can't drift.
Delete the leftover src/test/mappingHandlers.test.ts template stub (real tests are in tests/).

#1 and #2 before merge. Genuinely good work.