Mempool and policy: ranking notes

No P1 in this category: the two strongest items are relay/fee-estimation reliability bug fixes with clear downstream demand (#36182 is backport-targeted) but neither is a DoS/pinning or fund-safety failure, so both sit at the top of P2. The remaining work splits cleanly into policy/feature/test work with real use cases (P3) and narrow or internal changes that touch mempool code only incidentally (P4).

• #35581 states review is best focused on its base PR #35675, which extracts the BlockTemplateManager groundwork; the base should be reviewed first or the chain read together.
• #35286 (testsubmitpackage dry-run) and #36122 (BIP460 standardness) both target second-layer/collaborative-transaction use cases but are independent; #35286 addresses concrete reported demand (issue #32160) and should be reviewed first if a reviewer has time for only one.
• Two test-only PRs sit in different bands deliberately: #36214 exercises new, not-yet-stable cluster mempool invariants (P3) while #35882 only pins a reject string for long-enforced Taproot standardness (P4).
• #35557 is the only card noting an unanswered question about whether merged #35496 already covers its purpose; if so it becomes redundant, which would keep it at the bottom of P4.

This pass by claude-opus-5 on 2026-09-17 saw every PR in the category at once and checked the bands given to each PR alone against each other, ordered the PRs, and noted chains and overlaps. Back to the category.