feat(bundler): enhance error handling and nonce management in execute…

[onahprosper]

Description

This change hardens sponsored execution in app/lib/bundler/executeSponsored.ts so /api/bundler/execute-sponsored recovers when the sponsor wallet nonce is wrong or the RPC reports a stale nonce.

Previously, retry logic only treated an error as “nonce too low” if err.message contained that substring. Viem often puts the human-readable explanation in the top-level message (for example, nonce lower than the account’s current nonce) and puts nonce too low: next nonce … in details or nested RPC cause objects, so retries never ran and users saw 500s.

The implementation now collects text from the full viem error chain (including plain-object causes), matches several nonce-too-low phrasings, retries up to three times, chooses the next nonce with max(latest, pending) from eth_getTransactionCount, and when the RPC includes it, parses next nonce N from the error and uses the maximum of that and the RPC-derived nonce.

There are no API or contract changes. Behavior is limited to the bundler’s sponsor transaction path.

References

• Closes [BUG] Nonce too low on noblocks #466

Testing

How reviewers can test

1. Run Noblocks locally (pnpm dev) with a configured sponsor key and Base (or another EVM chain) RPC used by the bundler.
2. Trigger a flow that calls POST /api/bundler/execute-sponsored (EOA + EIP-7702 batch path).
3. Confirm the request succeeds when the chain nonce is correct; if a stale nonce scenario is hard to reproduce, temporarily misconfigure or simulate RPC lag only in a dev environment.

Not covered

• No new automated tests were added for this path; manual verification against a testnet or careful mainnet testing is the practical check.

Environment

• Developed with Node/pnpm and Next.js as in the Noblocks app; viem 2.x as used by the project.

• This change adds test coverage for new/changed/fixed functionality

Checklist

• I have added documentation and tests for new/changed functionality in this PR
• All active GitHub checks for tests, formatting, and security are passing
• The correct base branch is being used, if not main

By submitting a PR, I agree to Paycrest's Contributor Code of Conduct and Contribution Guide.

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 249b3cc8-2626-48b0-a1c8-0a9ae4adaccd

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch fix/bundler-execute-sponsored-nonce

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}