diff --git a/docs/autopsy.md b/docs/autopsy.md
index 0d8216530b..985a9fca48 100644
--- a/docs/autopsy.md
+++ b/docs/autopsy.md
@@ -4,36 +4,35 @@
 
 > A verdict here is a **candidate**, never a bank: the whole-binary byte-gate remains the sole arbiter (G3/P9).
 
-**Corpus:** 1752 record(s) · 1659 classified · status {'near': 1354, 'nobuild': 93, 'match': 305}
+**Corpus:** 1699 record(s) · 1610 classified · status {'near': 1295, 'nobuild': 89, 'match': 315}
 
 
 ## Buckets — what tool each failure actually wants
 
 | bucket | fns | reach-weighted | meaning |
 |--------|----:|---------------:|---------|
-| redraft | 699 | 2162 | the stored draft is NOT this function — un-attempted work misfiled as a near-miss; route to a fresh crack, not to a wall |
-| structural | 578 | 7761 | local mutation CANNOT introduce it — wants a C-level idiom, not CPU |
-| integration | 305 | 2303 | byte-correct standalone; blocked on plumbing (§58/§59 reconcile ladder) |
-| permuter | 75 | 740 | local mutation CAN reach it — permuter fuel (a plateau here is a **missing-transform** candidate) |
-| unknown | 2 | 2 | the residue the LLM tier is for |
+| redraft | 707 | 1638 | the stored draft is NOT this function — un-attempted work misfiled as a near-miss; route to a fresh crack, not to a wall |
+| structural | 528 | 5067 | local mutation CANNOT introduce it — wants a C-level idiom, not CPU |
+| integration | 315 | 2454 | byte-correct standalone; blocked on plumbing (§58/§59 reconcile ladder) |
+| permuter | 57 | 855 | local mutation CAN reach it — permuter fuel (a plateau here is a **missing-transform** candidate) |
+| unknown | 3 | 140 | the residue the LLM tier is for |
 
 ## Classes
 
 | class | fns | reach-weighted | bucket | permuter profile |
 |-------|----:|---------------:|--------|------------------|
-| SIZE-MISMATCH | 699 | 2162 | redraft | — |
-| LENGTH-DRIFT | 394 | 6380 | structural | — |
-| MATCH | 305 | 2303 | integration | — |
-| OPCODE-MIXED | 80 | 878 | structural | — |
-| WIDTH | 44 | 310 | structural | — |
+| SIZE-MISMATCH | 707 | 1638 | redraft | — |
+| LENGTH-DRIFT | 353 | 4223 | structural | — |
+| MATCH | 315 | 2454 | integration | — |
+| OPCODE-MIXED | 79 | 478 | structural | — |
+| WIDTH | 46 | 449 | structural | — |
 | IMM-OFFSET | 38 | 38 | structural | — |
-| DELAY-SLOT | 23 | 289 | permuter | schedule |
-| BRANCH-POLARITY | 22 | 155 | structural | — |
-| REGALLOC-PERM | 18 | 151 | permuter | regalloc |
-| ADDRESSING | 17 | 150 | permuter | cse |
-| IMM-VALUE | 10 | 143 | permuter | cse |
-| SCHEDULE-REORDER | 7 | 7 | permuter | schedule |
-| MIXED | 2 | 2 | unknown | — |
+| BRANCH-POLARITY | 21 | 21 | structural | — |
+| ADDRESSING | 19 | 418 | permuter | cse |
+| DELAY-SLOT | 18 | 284 | permuter | schedule |
+| REGALLOC-PERM | 7 | 7 | permuter | regalloc |
+| IMM-VALUE | 4 | 4 | permuter | cse |
+| MIXED | 3 | 140 | unknown | — |
 
 ## Top clusters (the batch-autopsy work items)
 
@@ -41,31 +40,35 @@ One representative per signature diagnoses the whole cluster — that is what ma
 
 | # | signature | fns | reach-weighted | example |
 |--:|-----------|----:|---------------:|---------|
-| 1 | `SIZE-MISMATCH/short` | 604 | 1801 | func_8017E918 (close=4, 2 ins) |
-| 2 | `MATCH` | 305 | 2303 | func_80168B70 (close=0, 27 ins) |
-| 3 | `LENGTH-DRIFT/-1?` | 116 | 1712 | func_8018312C (close=5, 7 ins) |
-| 4 | `SIZE-MISMATCH/long` | 95 | 361 | func_8017E06C (close=6, 9 ins) |
-| 5 | `LENGTH-DRIFT/-2?` | 70 | 1268 | func_80190480 (close=3, 4 ins) |
-| 6 | `OPCODE-MIXED` | 67 | 599 | func_80161D20 (close=1, 14 ins) |
-| 7 | `LENGTH-DRIFT/1?` | 63 | 595 | func_80187074 (close=4, 7 ins) |
-| 8 | `LENGTH-DRIFT/2?` | 51 | 450 | func_80181B8C (close=5, 11 ins) |
+| 1 | `SIZE-MISMATCH/short` | 614 | 1545 | func_8017FF90 (close=4, 2 ins) |
+| 2 | `MATCH` | 315 | 2454 | func_80168B70 (close=0, 27 ins) |
+| 3 | `LENGTH-DRIFT/-1?` | 112 | 1047 | func_8018312C (close=5, 7 ins) |
+| 4 | `SIZE-MISMATCH/long` | 93 | 93 | func_8017E06C (close=6, 9 ins) |
+| 5 | `LENGTH-DRIFT/-2?` | 70 | 1272 | func_80190480 (close=3, 4 ins) |
+| 6 | `OPCODE-MIXED` | 67 | 466 | func_80161D20 (close=1, 14 ins) |
+| 7 | `LENGTH-DRIFT/1?` | 62 | 461 | func_80187074 (close=4, 7 ins) |
+| 8 | `LENGTH-DRIFT/2?` | 50 | 316 | func_80181B8C (close=5, 11 ins) |
 | 9 | `LENGTH-DRIFT/-7` | 32 | 32 | func_80184590 (close=9, 2 ins) |
-| 10 | `LENGTH-DRIFT/+1` | 21 | 21 | func_8017DCE4 (close=3, 8 ins) |
-| 11 | `IMM-OFFSET/1` | 17 | 17 | func_801827C4 (close=1, 13 ins) |
-| 12 | `DELAY-SLOT/1` | 13 | 146 | func_8016BA68 (close=1, 29 ins) |
-| 13 | `BRANCH-POLARITY/bnez!=beqz` | 11 | 144 | func_80185E38 (close=1, 12 ins) |
-| 14 | `IMM-VALUE` | 10 | 143 | func_8018A640 (close=2, 4 ins) |
-| 15 | `WIDTH/lhu!=lh` | 9 | 9 | func_80142A10 (close=1, 28 ins) |
-| 16 | `LENGTH-DRIFT/-3?` | 8 | 1072 | func_8012CC88 (close=13, 102 ins) |
-| 17 | `OPCODE-MIXED/addressing,width` | 8 | 274 | func_8017F114 (close=5, 75 ins) |
-| 18 | `DELAY-SLOT/2` | 8 | 141 | func_80158638 (close=2, 87 ins) |
+| 10 | `IMM-OFFSET/1` | 17 | 17 | func_801827C4 (close=1, 13 ins) |
+| 11 | `DELAY-SLOT/1` | 10 | 143 | func_8016BA68 (close=1, 29 ins) |
+| 12 | `BRANCH-POLARITY/bnez!=beqz` | 10 | 10 | func_80185E38 (close=1, 12 ins) |
+| 13 | `WIDTH/lhu!=lh` | 9 | 9 | func_80142A10 (close=1, 28 ins) |
+| 14 | `BRANCH-POLARITY/beqz!=bnez` | 8 | 8 | func_8017FDB0 (close=1, 12 ins) |
+| 15 | `OPCODE-MIXED/addressing,width` | 6 | 6 | func_8017F114 (close=5, 75 ins) |
+| 16 | `DELAY-SLOT/2` | 6 | 139 | func_80158638 (close=2, 87 ins) |
+| 17 | `ADDRESSING/lui!=addiu` | 6 | 139 | func_8012A328 (close=3, 60 ins) |
+| 18 | `WIDTH/sh!=lw` | 5 | 5 | func_8018AFDC (close=3, 8 ins) |
+| 19 | `WIDTH/lw!=lhu` | 4 | 270 | func_8015FBE0 (close=1, 58 ins) |
+| 20 | `LENGTH-DRIFT/+1` | 4 | 4 | func_8017DCE4 (close=3, 8 ins) |
 
 ## Second oracle — deterministic verdict vs the recorded label
 
 The recorded `klass` is what a wave agent typed months ago; the verdict is derived from today's bytes. Neither is authoritative — **the disagreement is the finding**, and it is exactly the `mis-classified` bucket (a wrongly-labelled residual sends the directed search at the wrong levers, so it plateaus and then reads as a wall).
 
-- labelled records comparable: **4** · agree **3** · disagree **1**
+- labelled records comparable: **4** · agree **1** · disagree **3**
 
 | fn | recorded label | label routes to | measured class | measured routes to |
 |----|----------------|-----------------|----------------|--------------------|
+| func_80140D68 | `schedule` | schedule | ADDRESSING | cse |
+| func_8012A328 | `schedule` | schedule | ADDRESSING | cse |
 | func_801549F8 | `schedule` | schedule | ADDRESSING | cse |
diff --git a/phase-ends/CURRENT_PHASE.md b/phase-ends/CURRENT_PHASE.md
index 486ee04df9..79cf89c6b6 100644
--- a/phase-ends/CURRENT_PHASE.md
+++ b/phase-ends/CURRENT_PHASE.md
@@ -4507,3 +4507,45 @@ resolves. T0.1/T0.3 only read sigs/configs and write `.run/` + `docs/` ⇒ safe
   **⇒ The single highest-value backlog action is to RUN `residual_class.py` over the 1,486 unlabelled
   entries** (deterministic, LLM-free, reads the bytes) so the 333 close=0 + 184 close-1-4 band can be
   routed to the right lever instead of grinding undirected. That is a zero-token tooling run.
+
+- **✅ 2026-07-26 (SESSION-20) — T0.3b AUTOPSY: every backlog residual RECOMPUTED from the bytes
+  (1,699 rows, `-j 12`, zero agent tokens). Three findings, one of them a large honesty correction.**
+  `autopsy.py collect` re-derives each residual through the validated `match_one` path, deriving the
+  asm subdir AND the `-O0` flag from `corpus.py` (both are documented silent-artefact generators).
+  **R34 cross-check PASSED: closeness agreed with `masked_diff.structured_diff` on all 1,610 built
+  rows, 0 classifier errors.** 89 `nobuild` rows REPORTED, not silently dropped (R32).
+  **THE BUCKETS (reach-weighted in parens):**
+  | bucket | fns | meaning |
+  |---|---|---|
+  | **redraft** | **707** (1,638) | *the stored draft is NOT this function* — un-attempted work MISFILED as a near-miss |
+  | **structural** | **528** (5,067) | local mutation cannot introduce it — wants a C-level idiom, not CPU |
+  | **integration** | **315** (2,454) | **byte-correct standalone; blocked on PLUMBING** |
+  | permuter | **57** (855) | local mutation CAN reach it — the only true permuter fuel |
+  | unknown | 3 (140) | the residue the LLM tier is for |
+  **① THE HONESTY CORRECTION — 707 of 1,610 (44%) of the "near-miss backlog" ARE NOT NEAR-MISSES.**
+  Class `SIZE-MISMATCH` (614 `short` + 93 `long`): the stored best-draft is a PARTIAL — an agent's
+  incomplete attempt logged with a closeness score, exactly like `func_80183814`'s 666-of-5,122 recon.
+  **`docs/backlog.md` — "the phase's other durable deliverable" (PhaseEnd 21) — has been overstating
+  readiness by ~44% ever since.** These route to a FRESH CRACK, not to a wall and not to the permuter.
+  **② THE ACTIONABLE POOL — 315 entries are `match_one` MATCH RIGHT NOW**, blocked only on the
+  §58/§59 reconcile ladder. Recomputing beat trusting the stored label exactly as the tool's docstring
+  argued: the tree has moved (type-lifts, shared-header work) since these were logged. Top by value:
+  `func_80174CB0` (123×134 = 16,482) · `func_801463A0` (101×134 = 13,534) · `func_80173A60` (59×134) ·
+  `func_80176144` (53×138) · `func_8012F49C` (51×138); **total ≈108,959 gain-ins** where reach is
+  known. Spread across ~40 binaries (35 in ov_SC01_077). **⚠️ §52b still applies — ~half of close=0
+  drafts fail the whole-binary gate — so this is 315 CANDIDATES, not 315 banks.**
+  **③ THE PERMUTER BUCKET IS 57.** Task-13B measured 75/972 (7.7%) permuter-shaped and concluded
+  *"the problem was TARGETING, not a missing transform"*; the fresh recompute says **57 of 1,610
+  (3.5%)**. **Extending the mutation set is confirmed NOT the big lever** — it is a small, real, and
+  now precisely-bounded one.
+  **④ R34 EARNS ITS KEEP AGAIN — 3 of 4 comparable labels DISAGREE with the measurement:**
+  `func_80140D68`, `func_8012A328`, `func_801549F8` were all recorded **schedule** (→ schedule levers)
+  but measure **ADDRESSING** (→ cse levers). **The grinder was being pointed at the wrong levers for
+  these**, which is precisely the `mis-classified` bucket the autopsy exists to find.
+  **⇒ STRATEGIC CONVERGENCE (the answer to "what one thing unlocks the most"):** T0.2's gate probe
+  (8/8 failures PLUMBING, 0 walls) and T0.3b's autopsy (315 integration-class) **independently point
+  at the SAME lever — the declaration/integration reconcile ladder.** It is worth the 224,410-ins FREE
+  pool *and* ~108,959 backlog gain-ins. Two candidate keys were eliminated today (`--fix-def-sig`
+  regresses; `--normalize-self-decls` 0/123); the untried ones are **`canon_sig_reconcile` v3.2** and
+  **`reconcile_decls.py`** (the DATA-symbol analog — and one of T0.2's two failure texts was a DATA
+  symbol, `conflicting types for 'D_800A651C'`, which neither tested flag targets).