diff --git a/alkstore-contract-suite/src/lib.rs b/alkstore-contract-suite/src/lib.rs
index 689ee0e..251b8bd 100644
--- a/alkstore-contract-suite/src/lib.rs
+++ b/alkstore-contract-suite/src/lib.rs
@@ -38,9 +38,9 @@ pub mod version_stamp;
 
 pub use factory::StoreFactory;
 pub use properties::{
-    concurrent_try_lock_loser_is_a_value, duration_refusal_on_non_positive_ttl,
-    enqueue_opts_resolution, extent_clamp_semantics, in_tx_reads_see_own_writes,
-    job_handle_validity_predicate, lock_ttl_expiry_and_reacquisition,
+    backoff_curve_equivalence, concurrent_try_lock_loser_is_a_value,
+    duration_refusal_on_non_positive_ttl, enqueue_opts_resolution, extent_clamp_semantics,
+    in_tx_reads_see_own_writes, job_handle_validity_predicate, lock_ttl_expiry_and_reacquisition,
     name_validation_rejects_empty_and_reserved, outbox_enqueue_tx_commit_atomicity,
     payload_round_trip_stores_exact_encoding, payload_too_large_never_produced_on_sqlite,
     payload_too_large_produced_on_pg, publish_with_key_tx_commit_atomicity,
diff --git a/alkstore-contract-suite/src/properties.rs b/alkstore-contract-suite/src/properties.rs
index d2b8a55..d9fb8ec 100644
--- a/alkstore-contract-suite/src/properties.rs
+++ b/alkstore-contract-suite/src/properties.rs
@@ -894,15 +894,22 @@ pub async fn in_tx_reads_see_own_writes(factory: &dyn StoreFactory) {
 /// `EnqueueOpts::max_attempts` override riding over them — `get_job`
 /// shows resolved values, never the raw opts.
 ///
-/// The `run_at`-alone-literal and neither-field-resolves-to-now legs
-/// are wall-clock-adjacent; they pin with wave 5's equivalence row
-/// (this row's precedence/resolution legs are the engine-agnostic
-/// claim).
+/// Wave-5 completion — the two clock-adjacent legs the earlier note
+/// deferred pin here: the `run_at`-alone literal leg (a future
+/// `run_at` is the row's ready time *verbatim* — no arithmetic, a
+/// deterministic equality; a past `run_at` is ready *now* — the
+/// boundary-total rule, claimable through the `run_at <= now`
+/// predicate within a bounded wait) and the neither-field leg (ready
+/// now at the enqueue instant — tolerance-bounded proximity of the
+/// stamp to the clock read, never a tight timing assert).
 ///
-/// Contract stamp: ADR-020 §1 (delay wins over `run_at`); §2
+/// Contract stamp: ADR-020 §1 (delay wins over `run_at`; `run_at`
+/// alone is literal; neither field resolves to now); §2
 /// (relative `expires` → absolute row value); §3/§3a (the derived
 /// default stamps — plain queue 300/3/5/none, per-job override);
-/// ADR-012 §2 (engine-side arithmetic pinned equivalent).
+/// ADR-023 §2 (the boundary-total rule — a past ready time is ready
+/// now, never rejected); ADR-012 §2 (engine-side arithmetic pinned
+/// equivalent).
 pub async fn enqueue_opts_resolution(factory: &dyn StoreFactory) {
     let store = factory.open().await.expect("factory opens a store");
 
@@ -993,6 +1000,259 @@ pub async fn enqueue_opts_resolution(factory: &dyn StoreFactory) {
         "the per-job override rides over the derived defaults, others untouched"
     );
 
+    // Wave-5 clock legs (the run_at-alone and neither-field pins).
+    // Future run_at alone: the ready time is the literal — stored
+    // verbatim, no arithmetic, no clock read involved (a deterministic
+    // equality, not a timing assert).
+    let literal = 1_800_000_000;
+    let literal_id = queue
+        .enqueue(
+            json!({"leg": "run-at-literal"}),
+            EnqueueOpts {
+                run_at: Some(literal),
+                ..Default::default()
+            },
+        )
+        .await
+        .unwrap();
+    let literal_job = queue.get_job(literal_id).await.unwrap().unwrap();
+    assert_eq!(
+        literal_job.run_at, literal,
+        "run_at alone is the row's ready time verbatim (delay absent)"
+    );
+
+    // Past run_at alone: ready now (the boundary-total rule) —
+    // claimable within a bounded wait. The claim ordering (run_at ASC)
+    // hands it out ahead of the every same-now row above, so the
+    // observation is unambiguous even against the row's other legs.
+    let past = unix_now() - 30;
+    let past_id = queue
+        .enqueue(
+            json!({"leg": "run-at-past"}),
+            EnqueueOpts {
+                run_at: Some(past),
+                ..Default::default()
+            },
+        )
+        .await
+        .unwrap();
+    assert_eq!(
+        queue.get_job(past_id).await.unwrap().unwrap().run_at,
+        past,
+        "a past run_at stores the literal too — ready-now is the claim semantics"
+    );
+    let claimed_past = tokio::time::timeout(std::time::Duration::from_secs(5), async {
+        loop {
+            if let Some(h) = queue.claim_one("wave5-past").await.unwrap() {
+                let id = h.job().id;
+                if id == past_id {
+                    break id;
+                }
+            }
+            tokio::time::sleep(std::time::Duration::from_millis(50)).await;
+        }
+    })
+    .await
+    .expect("a past run_at resolves ready-now: claimable within the bounded wait");
+    assert_eq!(
+        claimed_past, past_id,
+        "the past-run_at row is the ready-now claim (ordering: run_at ASC)"
+    );
+
+    // Neither field: ready now at the enqueue instant — the stamp sits
+    // close to the clock read (tolerance-bounded proximity, never a
+    // tight timing assert).
+    let neither_id = queue
+        .enqueue(json!({"leg": "neither"}), EnqueueOpts::default())
+        .await
+        .unwrap();
+    let now = unix_now();
+    assert!(
+        queue
+            .get_job(neither_id)
+            .await
+            .unwrap()
+            .unwrap()
+            .run_at
+            .abs_diff(now)
+            <= 5,
+        "neither field resolves to now: the stamp is within the tolerance \
+         of the clock read"
+    );
+
+    factory.teardown().await.expect("factory teardown");
+}
+
+/// The backoff row's bounded-wait claim: poll `claim_one` on `queue`
+/// until `id` comes out and return that claim's handle (claim
+/// ordering keeps single-job queues unambiguous; the deadline is the
+/// safety bound — the assertions are on state outcomes, and the claim
+/// is consumed here, not re-taken afterwards).
+async fn await_claimable(
+    queue: &dyn alkstore::Queue,
+    id: i64,
+    max_wait: std::time::Duration,
+    label: &str,
+) -> Box<dyn alkstore::JobHandle> {
+    let deadline = std::time::Instant::now() + max_wait;
+    loop {
+        match queue.claim_one("backoff-w").await.unwrap() {
+            Some(h) if h.job().id == id => return h,
+            _ => {}
+        }
+        assert!(
+            std::time::Instant::now() < deadline,
+            "{label}: the row never became ready within the bounded wait"
+        );
+        tokio::time::sleep(std::time::Duration::from_millis(100)).await;
+    }
+}
+
+/// **Backoff-curve equivalence across engines** — the equal-jitter
+/// exponential curve (ADR-010 §3): `retry(err, None)` computes
+/// `delay ∈ [base·2^(a−1)/2, base·2^(a−1)]` from the job's stamped
+/// `backoff_base_s` (the attempt index `a` the row's post-claim
+/// `attempts`), capped at 1 hour — *the range, not a jitter label, is
+/// the definition*, and identical bounds across engines is the
+/// equivalence pin (ADR-012 §2: engine-side arithmetic, suite-pinned
+/// identical outputs). The row drives claim → `retry(err, None)`
+/// cycles on a queue with `backoff_base_s = 2` (integer-second
+/// resolution: the ranges [1,2], [2,4], [4,8] land across the first
+/// three attempts), reading each computed delay back through
+/// `get_job`'s resolved `run_at` minus a clock read taken before the
+/// retry, asserting range membership per attempt — with a
+/// one-second straddle tolerance on the upper side only (the retry's
+/// internal clock read and the suite's may cross a wall second; the
+/// lower side is race-free, the retry's clock read cannot precede the
+/// suite's). The inter-attempt waits are tolerance-bounded and
+/// sequential (the row waits for each `run_at` to pass through
+/// bounded-wait claims).
+///
+/// The explicit-delay override legs: `retry(err, Some(d))` is honored
+/// verbatim — `Some(5)` lands the row ready at the retry instant plus
+/// exactly 5 s (the same straddle tolerance), and `Some(0)` resolves
+/// ready-now (the boundary-total rule: a zero delay is claimable at
+/// the retry's own second, within a bounded wait).
+///
+/// Cap-leg disposition: the 1-hour cap is not suite-pinnable (reaching
+/// capped attempts requires waiting out minute-scale delays); it stays
+/// pinned engine-side — both engines' unit tests pin the curve against
+/// the ADR text including the cap.
+///
+/// Contract stamp: ADR-010 §3 (the equal-jitter exponential, the range
+/// as definition, the 1-hour cap); ADR-020 §1 (the explicit-delay
+/// override resolves to the row's ready time — `Some(0)` ready now);
+/// ADR-012 §2 (engine-side arithmetic pinned identical); ADR-023 §2
+/// (the boundary-total rule — a zero delay is ready now, never
+/// rejected).
+pub async fn backoff_curve_equivalence(factory: &dyn StoreFactory) {
+    let store = factory.open().await.expect("factory opens a store");
+    let queue = store
+        .queue(
+            "curve",
+            alkstore::QueueOpts {
+                max_attempts: 7,
+                backoff_base_s: 2,
+                ..Default::default()
+            },
+        )
+        .await
+        .unwrap();
+    let id = queue
+        .enqueue(json!({"leg": "curve"}), EnqueueOpts::default())
+        .await
+        .unwrap();
+
+    // Three curve cycles at the first three attempts (base 2 s:
+    // ranges [1,2], [2,4], [4,8] — the integer-second row resolution
+    // kept honest by the 1-second straddle tolerance on the upper
+    // side). The now-read before the retry makes the lower bound
+    // race-free: the retry's internal clock read cannot precede it, so
+    // the observed value over-reads the delay, never under-reads.
+    for (attempt, lower, upper) in [(1i64, 1i64, 2i64), (2, 2, 4), (3, 4, 8)] {
+        let held = await_claimable(
+            queue.as_ref(),
+            id,
+            std::time::Duration::from_secs(upper as u64 + 10),
+            &format!("attempt {attempt}"),
+        )
+        .await;
+        assert_eq!(
+            held.job().attempts,
+            attempt,
+            "the attempt index is the claim count"
+        );
+        let now = unix_now();
+        assert!(
+            held.retry(Some("boom".to_string()), None)
+                .await
+                .expect("an in-window retry lands")
+        );
+        let job = queue
+            .get_job(id)
+            .await
+            .unwrap()
+            .expect("the retried row lives");
+        assert_eq!(
+            job.state,
+            JobState::Pending,
+            "the retry left the row pending"
+        );
+        let observed = job.run_at - now;
+        assert!(
+            (lower..=upper + 1).contains(&observed),
+            "attempt {attempt}: the computed delay must land in \
+             [{lower}, {upper}] (one-second straddle tolerance), got {observed}"
+        );
+    }
+
+    // Explicit-delay override: Some(5) is honored — ready at the retry
+    // instant plus exactly 5 s (the same straddle tolerance), not the
+    // attempt-4 curve range [8, 16].
+    let held = await_claimable(
+        queue.as_ref(),
+        id,
+        std::time::Duration::from_secs(10),
+        "Some(5) leg",
+    )
+    .await;
+    let now = unix_now();
+    assert!(
+        held.retry(Some("boom".to_string()), Some(5))
+            .await
+            .expect("an in-window retry lands")
+    );
+    let job = queue.get_job(id).await.unwrap().unwrap();
+    let observed = job.run_at - now;
+    assert!(
+        (5..=6).contains(&observed),
+        "retry(err, Some(5)) is honored verbatim (± one second straddle), \
+         got {observed}"
+    );
+
+    // Some(0): the boundary-total rule — ready now. The retry's own
+    // second already satisfies the claim predicate, so the bounded
+    // wait is short and the row lands promptly.
+    let held = await_claimable(
+        queue.as_ref(),
+        id,
+        std::time::Duration::from_secs(8),
+        "Some(0) leg",
+    )
+    .await;
+    assert!(
+        held.retry(Some("boom".to_string()), Some(0))
+            .await
+            .expect("an in-window retry lands")
+    );
+    await_claimable(
+        queue.as_ref(),
+        id,
+        std::time::Duration::from_secs(5),
+        "Some(0) resolves ready-now (the boundary-total rule)",
+    )
+    .await;
+
     factory.teardown().await.expect("factory teardown");
 }
 
diff --git a/alkstore-postgres/tests/contract_suite.rs b/alkstore-postgres/tests/contract_suite.rs
index e188f4a..f6fd9a9 100644
--- a/alkstore-postgres/tests/contract_suite.rs
+++ b/alkstore-postgres/tests/contract_suite.rs
@@ -163,16 +163,16 @@ async fn harness_ready() -> bool {
 
 mod rows {
     use alkstore_contract_suite::properties::{
-        concurrent_try_lock_loser_is_a_value, drop_rollback_leaves_no_ghosts,
-        duration_refusal_on_non_positive_ttl, enqueue_opts_resolution, extent_clamp_semantics,
-        in_tx_reads_see_own_writes, job_handle_validity_predicate,
-        lock_ttl_expiry_and_reacquisition, name_validation_rejects_empty_and_reserved,
-        outbox_enqueue_tx_commit_atomicity, payload_round_trip_stores_exact_encoding,
-        payload_too_large_produced_on_pg, publish_with_key_tx_commit_atomicity,
-        queue_depth_reclaim_and_dead_letter, receiver_close_and_save_arms,
-        scheduler_boundary_fires, scheduler_bounded_catchup, scheduler_leadership_discipline,
-        stream_ordering_equivalence, sweep_no_stranded_rows, trim_to_semantics,
-        wake_receiver_shapes, with_tx_panicking_closure_rolls_back,
+        backoff_curve_equivalence, concurrent_try_lock_loser_is_a_value,
+        drop_rollback_leaves_no_ghosts, duration_refusal_on_non_positive_ttl,
+        enqueue_opts_resolution, extent_clamp_semantics, in_tx_reads_see_own_writes,
+        job_handle_validity_predicate, lock_ttl_expiry_and_reacquisition,
+        name_validation_rejects_empty_and_reserved, outbox_enqueue_tx_commit_atomicity,
+        payload_round_trip_stores_exact_encoding, payload_too_large_produced_on_pg,
+        publish_with_key_tx_commit_atomicity, queue_depth_reclaim_and_dead_letter,
+        receiver_close_and_save_arms, scheduler_boundary_fires, scheduler_bounded_catchup,
+        scheduler_leadership_discipline, stream_ordering_equivalence, sweep_no_stranded_rows,
+        trim_to_semantics, wake_receiver_shapes, with_tx_panicking_closure_rolls_back,
     };
 
     use super::PgFactory;
@@ -243,6 +243,11 @@ mod rows {
         enqueue_opts_resolution,
         "row-opts-resolution"
     );
+    harness_row!(
+        row_backoff_curve_equivalence,
+        backoff_curve_equivalence,
+        "row-backoff-curve"
+    );
     harness_row!(
         row_receiver_close_and_save_arms,
         receiver_close_and_save_arms,
diff --git a/alkstore-sqlite/tests/contract_suite.rs b/alkstore-sqlite/tests/contract_suite.rs
index 6f74d16..bd3ed6c 100644
--- a/alkstore-sqlite/tests/contract_suite.rs
+++ b/alkstore-sqlite/tests/contract_suite.rs
@@ -67,16 +67,16 @@ impl StoreFactory for SqliteFactory {
 
 mod rows {
     use alkstore_contract_suite::properties::{
-        concurrent_try_lock_loser_is_a_value, drop_rollback_leaves_no_ghosts,
-        duration_refusal_on_non_positive_ttl, enqueue_opts_resolution, extent_clamp_semantics,
-        in_tx_reads_see_own_writes, job_handle_validity_predicate,
-        lock_ttl_expiry_and_reacquisition, name_validation_rejects_empty_and_reserved,
-        outbox_enqueue_tx_commit_atomicity, payload_round_trip_stores_exact_encoding,
-        payload_too_large_never_produced_on_sqlite, publish_with_key_tx_commit_atomicity,
-        queue_depth_reclaim_and_dead_letter, receiver_close_and_save_arms,
-        scheduler_boundary_fires, scheduler_bounded_catchup, scheduler_leadership_discipline,
-        stream_ordering_equivalence, sweep_no_stranded_rows, trim_to_semantics,
-        wake_receiver_shapes, with_tx_panicking_closure_rolls_back,
+        backoff_curve_equivalence, concurrent_try_lock_loser_is_a_value,
+        drop_rollback_leaves_no_ghosts, duration_refusal_on_non_positive_ttl,
+        enqueue_opts_resolution, extent_clamp_semantics, in_tx_reads_see_own_writes,
+        job_handle_validity_predicate, lock_ttl_expiry_and_reacquisition,
+        name_validation_rejects_empty_and_reserved, outbox_enqueue_tx_commit_atomicity,
+        payload_round_trip_stores_exact_encoding, payload_too_large_never_produced_on_sqlite,
+        publish_with_key_tx_commit_atomicity, queue_depth_reclaim_and_dead_letter,
+        receiver_close_and_save_arms, scheduler_boundary_fires, scheduler_bounded_catchup,
+        scheduler_leadership_discipline, stream_ordering_equivalence, sweep_no_stranded_rows,
+        trim_to_semantics, wake_receiver_shapes, with_tx_panicking_closure_rolls_back,
     };
 
     use super::SqliteFactory;
@@ -135,6 +135,11 @@ mod rows {
         enqueue_opts_resolution(&factory("row-opts-resolution")).await;
     }
 
+    #[tokio::test(flavor = "multi_thread")]
+    async fn row_backoff_curve_equivalence() {
+        backoff_curve_equivalence(&factory("row-backoff-curve")).await;
+    }
+
     #[tokio::test(flavor = "multi_thread")]
     async fn row_receiver_close_and_save_arms() {
         receiver_close_and_save_arms(&factory("row-receiver-arms")).await;
diff --git a/tasks/suite-opts-backoff-rows.md b/tasks/suite-opts-backoff-rows.md
index 9d3ac24..d252622 100644
--- a/tasks/suite-opts-backoff-rows.md
+++ b/tasks/suite-opts-backoff-rows.md
@@ -1,7 +1,7 @@
 ---
 id: suite-opts-backoff-rows
 name: Contract-suite rows — backoff-curve equivalence + the deferred enqueue-opts clock legs
-status: pending
+status: completed
 depends_on: []
 scope: narrow
 risk: low
@@ -76,6 +76,80 @@ Work:
 
 > To be filled by implementation agent
 
+- **The 1-hour cap leg is not suite-pinned** (as pre-decided): reaching
+  capped attempts requires waiting out minute-scale delays. The cap
+  stays pinned engine-side — both engines' unit tests pin the curve
+  against the ADR text including the cap (SQLite's
+  `backoff_curve_lands_in_the_pinned_range` and the pg twin, verified
+  in the wave-3/4 rounds); the disposition is recorded in the row's
+  doc comment for the review gate.
+- **Observation tolerance convention for the curve row**: the
+  computed delay is read back as `get_job`'s `run_at` minus a clock
+  read taken *before* the `retry(err, None)` call. That direction is
+  deliberate and race-free on the lower side: the retry's internal
+  clock read (SQLite's `unixepoch()` in-statement, pg's Rust-side
+  engine read) cannot precede the suite's, so the observed value
+  *over-reads* the delay, never under-reads — the pinned lower bound
+  is exact. The upper bound carries a one-second straddle tolerance
+  (retry commit and suite read may cross a wall second, and the
+  engines' stamps are integer seconds, ADR-010 §3's integerized
+  resolution). Asserted as a `(lower..=upper+1)` range membership; the
+  *identical* assertion on both engines is the equivalence pin
+  (ADR-012 §2's identical-outputs rule) — the row body is shared, so
+  the bounds are identical by construction.
+- **Base 2 s kept the integer-second row resolution honest** per the
+  task text: ranges [1,2], [2,4], [4,8] across attempts 1–3 (the a=1
+  pin is inherently weak at one-second resolution — lower bound 1 with
+  upper ±1 — but the attempt-2/3 ranges are genuinely discriminating,
+  and the explicit-delay legs pin the exact-literal resolution).
+- **Past-run_at leg stores the literal too**: the pin on the past leg
+  is the *claim semantics* (ready-now through the `run_at <= now`
+  predicate, bounded-wait claimable) and the stored literal (no
+  arithmetic on any form) — "resolves to ready now" is the claim-
+  predicate reading of the boundary-total rule (ADR-023 §2), not a
+  clamping of the stored value; both engines store verbatim
+  (`resolution.rs` twins verified).
+- **Claim-polling observation posture**: both rows' ready-now waits
+  are bounded `claim_one` poll loops (100/50 ms cadence), deadline
+  asserts only — no timing-value assertions, single-store sequential
+  drive within the module doc's determinism posture.
+- **One unreproducibility note on the helper**: the first draft made
+  `await_claimable` re-claim after the bounded wait, which found
+  nothing (the bounded wait consumed the claim); the helper returns
+  the claimed handle. Found on the first SQLite run, no engine-side
+  hazard.
+- No ADR amendments needed; no divergences found (both engines'
+  arithmetic agreed out of the box — the equivalence pin held on the
+  first full run pairs).
+
 ## Summary
 
-> To be filled on completion
\ No newline at end of file
+> To be filled on completion
+
+`enqueue_opts_resolution` extended in place with the wave-5
+completion legs (its deferral note replaced by the completion
+statement, ADR-023 §2 stamp accumulated): the `run_at`-alone literal
+leg (future `run_at` = the row's ready time verbatim — a
+deterministic equality, no clock read involved; past `run_at` stored
+verbatim too and *claimable now* within a bounded-wait claim, the
+run_at-ASC ordering making the observation unambiguous among the
+row's other legs) and the neither-field leg (ready now at the
+enqueue instant, `abs_diff(now) <= 5` tolerance-bounded proximity —
+never a tight timing assert). New `backoff_curve_equivalence` row:
+claim → `retry(err, None)` cycles on a `backoff_base_s = 2` queue
+(ranges [1,2], [2,4], [4,8] per attempt, asserted as `(lower..=upper+1)`
+range membership of `run_at − now_pre` with the race-free lower
+bound and one-second straddle tolerance documented), reading each
+computed delay back through `get_job`'s resolved `run_at`; the
+explicit-delay override legs ride the same row (`Some(5)` honored
+verbatim in [5, 6]; `Some(0)` resolving ready-now per the
+boundary-total rule, claimed within a bounded wait). Stamped ADR-010
+§3, ADR-020 §1, ADR-012 §2, ADR-023 §2. Both rows wired into both
+engines' `contract_suite.rs` targets (SQLite `#[tokio::test]`, pg
+`harness_row!`) — 25 rows per column, up from 24. Verified: SQLite
+column green server-less (25/25 suite, 189 lib), pg column green vs
+the harness server (postgres/poc@:15432 — 25/25 suite, 121 lib + 9
+schema), 3 solo re-runs of each new/extended row per engine
+(determinism), server-less pg column skips cleanly (25 "passed" via
+the reachability gate), `cargo test -p alkstore-sqlite -p
+alkstore-postgres` green, clippy `-D warnings`, fmt clean.
\ No newline at end of file