fix[oz-retainer-07-m-02]: remove OOReporter runtime liveness cap

[md0x]

Audit finding

OpenZeppelin Retainer 07 identified the following issue:

M-02 — Raising minimumDisputeWindow Above a Registered Liveness Range Permanently Blocks a Request

• Severity: Medium
• Status: Open

OOReporter.registerRequest stores immutable liveness bounds that are validated against Managed OO configuration only at registration time. If minimumDisputeWindow later rises above the stored maximumLiveness, no value can satisfy both the reporter's registered range and Managed OO's current minimum. This can permanently block initialization and manual recovery for the affected request.

References: OpenZeppelin audit findings · FRO-96 · audited scope tag

Resolution

• Enforce the registered minimumLiveness as the reporter's runtime floor.
• Retain maximumLiveness unchanged in storage, return values, and events as an offchain initialization target rather than an onchain ceiling.
• Continue calling setCustomLiveness, so Managed OO enforces its current minimumDisputeWindow and technical maximum.
• Allow manual recovery above a stale registered maximum after an automatic re-request fails because it reused the previous liveness.
• Preserve registration validation, public ABI shapes, events, and custom-error signatures.

This is the runtime remediation specified in FRO-107. The separate request-manager override revalidation finding remains outside this PR.

Validation

• cd pm-v2-oo-reporter && forge fmt --check
• cd pm-v2-oo-reporter && forge test --match-path test/OOReporter.t.sol — 41 tests passed
• git diff --check

[Reinis-FRP]

looks good!

[chrismaree]

Reviewed against FRO-96 and FRO-107. The change removes only the stale registered maximum as a runtime ceiling, preserves the registered minimum and ABI/storage shapes, and continues delegating current protocol bounds to Managed OO. The configuration-drift and manual-recovery coverage matches the issue acceptance criteria.

---

Sent from Chris Codex Agent 🤖