ZARP v1 Retrospective — ZIP Audit Reimbursement Program (ZARP)

This post provides the closeout report for ZARP v1, the ZIP Audit Reimbursement Program approved under [TPP-3] in May 2025. ZARP was created to make high-quality security work the default standard for ZKsync protocol upgrades by reimbursing eligible audit-related costs for successfully executed ZKsync Improvement Proposals (ZIPs) during the 2025 calendar year.

Security is a core dependency for any protocol. ZARP exists to reduce friction for rigorous security reviews while preserving governance oversight, verifiability, and transparent onchain disbursement mechanics.

Program recap

ZARP v1 was approved as an annual program valued at $5,000,000 USD, denominated as 89,285,714 ZK using the reference pricing method specified in the original proposal. The program was implemented via two capped minters:

  • ZarpRetro: reimbursements for ZIPs approved by the Token Assembly between 1 January and 30 April 2025
  • ZarpMain: reimbursements for ZIPs executed between 1 May and 31 December 2025, via a per-ZIP child capped minter flow

Onchain contracts (ZARP v1)

  • ZarpMain: 0x51E818785dEa065D392ac21F04E9cac5B601Cfd8
  • ZarpRetro: 0x70F6998FC0c492d9DD08b1105259252329be9Db6

Summary outcome

ZARP v1 operated in line with the intended security and governance model:

  • ZarpRetro reimbursements were reviewed by the Security Council and aligned with the eligible scope defined in the proposal (third-party audits, formal verification, and code competitions).
  • ZarpMain maintained the intended onchain flow for future reimbursements (via child capped minters and a post-vote buffer). A portion of 2025 protocol security costs did not get reimbursed when ZIP-13 was submitted due to timing and process coupling between audit execution/invoicing and ZIP lifecycles.
  • This gap has been addressed through the governance-approved, one-time reconciliation mechanism included in ZARP v2 (details below), ensuring eligible 2025 security work can still be covered in a contained and transparent way.

Capped minter overview for ZARP v1

Capped minter Purpose Cap (ZK) Window Status
ZarpRetro Retro reimbursements (Q1 2025 ZIPs) 39,475,000 19 May 2025 – 31 Jan 2026 Reviewed by the ZKSC and approved by the Token Assembly
ZarpMain 2025 reimbursements (May–Dec 2025 ZIPs) 49,810,714 19 May 2025 – 31 Jan 2026 Unspent portion remained due to timing mismatch with ZIP-13.

Retro reimbursements (ZarpRetro)

Retro reimbursements claimed under ZarpRetro, as defined in the program documentation:

ZIP Amount claimed (USD)
ZIP-3 Protocol Defense 91,440
ZIP-6 Gateway Prep 1,490,540
ZIP-9 EVM Emulator 628,620
Total 2,210,600

Reimbursement categories included third-party audit work and competitive security reviews, consistent with the program’s stated eligibility. All reimbursements claimed from the ZarpRetro cap were reviewed and approved by the Security Council.

Main program reimbursements (ZarpMain)

ZarpMain was designed for reimbursements tied to ZIPs executed between May and December 2025 and required ZIP authors to complete key steps before onchain submission (including invoice submission for verification and deployment of a child capped minter with the relevant role-grant calldata).

Summary

  • The cap for ZarpMain was 49,810,714 ZK ($2.5m USD at 0.05).
  • A total of 10.6m ZK was minted from ZarpMain ($530k USD).
  • 39m ZK remained unminted on 31 January 2026 ($1.95m USD).

Reimbursements claimed under ZarpMain:

ZIP Amount claimed (USD) ZK Claimed Link Date CM Submitted
ZIP-11 Gateway Prep 270,286 5,405,720 ZK ZIP-11 child minter: 0x0455e47Ae27A20E026e69D69c4687d8e3F4ce635 26 September 2026
ZIP-12 EVM Emulator 260,000 5,200,000 ZK ZIP-12 child minter: 0xA790EF548B27aC62D36Cdc86979e8F606CC8850a 26 September 2025
Total 530,286

In practice, some protocol-critical security work in Q3 2025 did not align cleanly with the ZIP-specific timing assumptions in ZARP v1, particularly where audits and invoicing schedules did not match the expected claim setup window. This resulted in a portion of 2025 security costs not being reimbursed via ZarpMain during the v1 period, despite being security-relevant work for the protocol.

Relationship to ZARP v2 (2026 successor)

ZARP v2 has now been approved as the successor program for 2026. It preserves the same governance goals and security-first intent, while incorporating operational learnings from v1.

In particular, ZARP v2 includes:

  • A 2026 forward-looking capped minter for protocol-related security costs of $3m USD in ZK tokens (150m ZK); and
  • A one-time 2025 reconciliation capped minter to reimburse eligible 2025 protocol security audits that were not reimbursed under ZARP v1 due to timing misalignment between audit execution/invoicing and ZIP lifecycles, including ZIP-13-related security work, of $1.1m USD in ZK tokens (55m ZK).

This means just over half ($1.1m USD) of the “unspent” portion of ZARP v1 ($1.95m USD) was approved for claiming with the ZARP v2 Retro capped minter. The ZARP v2 reconciliation is explicitly governed, bounded, and documented onchain.

For reference, ZARP v2 capped minters are:

Notes and learnings

ZARP v1 demonstrated that governance-authorized, onchain reimbursement can support strong security standards with clear oversight and transparent funding mechanics. The primary learning from v1 was that protocol-critical security work is not always tightly coupled to discrete ZIP timelines or processes.

ZARP v2 responds directly to this by expanding eligibility beyond ZIP-only reimbursements to include protocol-critical security work executed under existing governance authorizations, while maintaining Security Council verification and transparency requirements, and by adding a one-time retroactive reconciliation path for 2025.

Conclusion

ZARP is a security enablement program. Its purpose is to make thorough third-party security review a default requirement for protocol change, while keeping reimbursement decisions verifiable, governed, and transparent. ZARP v1 achieved this objective within the constraints of its original ZIP-coupled design, and ZARP v2 builds on those foundations to better match how protocol security work is executed in practice.