diff --git a/active.md b/active.md
index 6d4cf305..fb25a165 100644
--- a/active.md
+++ b/active.md
@@ -13,36 +13,3 @@
- registered: 2026-08-19 by the wake_up session — the issuing session (claude/autofit-priors-messages-audit-ylvenv)
filed the prompt + issue but not this entry, tripping Lifecycle Drift on main.
- repos-none-claimed: this entry claims NO repos — one line deliberately, not 2-space bullets.
-
-## organ-repo-spelling-splits
-- issue: https://github.com/PyAutoLabs/PyAutoBrain/issues/287
-- issued: 2026-08-26
-- prompt: active/organ_repo_spellings_split_across_keys.md
-- status: ready-to-ship — implemented and pushed on both branches; no PR opened
- yet (PyAutoMind e160d40b, PyAutoBrain 6329db8). 515 Brain tests pass,
- repos_sync --check all 13 legs OK, Heart NOT EVALUATED (web-github).
-- worktree: n/a — web-github session; the session's own PyAutoBrain + PyAutoMind
- clones are the checkout, on the designated branch below (no
- ~/Code/PyAutoLabs-wt/ worktree exists in this environment).
-- repos:
- - PyAutoBrain: claude/organ-repo-spelling-splits-hbbms0
- - PyAutoMind: claude/organ-repo-spelling-splits-hbbms0
-- summary: |
- Infrastructure task (organism repos only). Fourth instance of the
- hand-maintained-aliases vs derived-target-set defect class (#267, #269);
- this one closes the class. Keying decision taken up front per the prompt:
- organs key PREFIXED (`pyautobrain`), extending #271's rule — canonical key
- is the bare package name where the repo ships one, the repo name where it
- does not. Fix derives the bare/prefixed/package alias join from the body
- map (new `package:` identity key in repos.yaml) and adds three coverage
- guards to test_policy_seams.py. Sweep also covers PyAutoScientist and
- pyautolabs.github.io, both split/unresolvable and not named in the prompt.
- Heart unreachable from this environment — ship gate leg 4 records NOT
- EVALUATED.
-
- DEVIATION from the answered scope question: `pyautolabs.github.io` was
- deliberately DE-REGISTERED rather than joined. Joining it means aliasing its
- truncated head `pyautolabs`, which is the ORG name — `@PyAutoLabs/PyAutoFit`
- would then resolve to the static site instead of PyAutoFit. The acceptance
- criterion's other branch ("or is deliberately not registered") covers it, and
- a guard pins both halves. Raised for review at ship time.
diff --git a/active/organ_repo_spellings_split_across_keys.md b/active/organ_repo_spellings_split_across_keys.md
deleted file mode 100644
index 45c40534..00000000
--- a/active/organ_repo_spellings_split_across_keys.md
+++ /dev/null
@@ -1,72 +0,0 @@
-# Organ repo spellings split across two normalised keys
-
-Type: bug
-Target: pyautobrain
-Repos:
-- PyAutoBrain
-Difficulty: large
-Autonomy: supervised
-Priority: normal
-Status: formalised
-Filed: 2026-08-24
-Issued: 2026-08-26
-
-Found while shipping the refactor witness-map audit (@PyAutoBrain#269 / PR #271).
-
-## The defect
-
-The bare spellings of the organ repos do not normalise to the same key as their
-repo spellings, so a policy map keyed on one is invisible to the other.
-
-@PyAutoBrain/agents/faculties/sizing/_sizing.py (`_target_sets.names_for`)
-registers BOTH spellings of every `PyAuto*` repo as known targets
-(`name.lower()` and `name.lower()[2:]`), but the `repo_aliases` table in
-@PyAutoBrain/config/policy.yaml only joins them for the libraries:
-
-| mention | normalises to | resolves? |
-|---|---|---|
-| `@PyAutoBrain` | `pyautobrain` | yes |
-| `@autobrain` | `autobrain` | no — a known target with nothing filed under it |
-
-Same for the bare Heart, Memory, Mind and Gut spellings. `@autohands` was joined
-in PR #271 because `extra_organism_targets` declares it explicitly and it
-reaches real code today; the rest are latent — no prompt in the Mind history
-uses them yet.
-
-## Decide before fixing
-
-This is NOT a blind alias sweep. Should organs key on the bare form
-(`autobrain`) or the prefixed form (`pyautobrain`)? `target_signals` keys the
-prefixed form (`pyautobrain`, `pyautomind`, `pyautoheart`, `pyautobuild`), and
-after #271 `test_witness` keys organs prefixed but libraries bare. Whichever way
-it goes has to hold across both maps at once, so answer the question first and
-let the edit follow.
-
-## The wider question this raises
-
-Three separate repos have now hit this same defect class — PyAutoNerves
-(#267), PyAutoCTI and PyAutoReduce (#269). The shared cause is that
-`repo_aliases` is HAND-MAINTAINED while the known-target set is DERIVED from the
-body map, so the two drift silently and the gap only ever surfaces as a
-wrong-but-plausible conductor message ("strengthen tests first" for a
-well-tested repo).
-
-Guard 2 added in #271 — a witness-map key must be what its repo normalises to —
-closes the witness-map instance only. An alias gap in a map with no coverage
-guard is still invisible. Consider deriving `repo_aliases` from the body map so
-the class closes rather than its fourth instance.
-
-## Notes on this prompt's header
-
-`pyauto-brain intake classify` scored this `bug` at high confidence (kept), but
-proposed `Target: autocti` — it read the prior-instance history above as the
-subject — and `Difficulty: too-large` (score 10), inflated by the repo names and
-the design-decision keywords. Corrected to `pyautobrain` / `large` on review.
-An alternative classification is `refactor`: the fix is behaviour-preserving for
-every spelling that resolves today. `bug` was kept because the Bug Agent's
-investigate-first strategy matches the "decide before fixing" requirement above.
-
-Acceptance: every spelling of every body-map repo that `_target_sets` registers
-as a known target resolves to a key the policy maps are actually filed under, or
-is deliberately not registered; a guard makes a future alias gap fail loudly
-rather than degrade a conductor's advice.
diff --git a/complete/2026/08/organ-repo-spelling-splits.md b/complete/2026/08/organ-repo-spelling-splits.md
new file mode 100644
index 00000000..d966109f
--- /dev/null
+++ b/complete/2026/08/organ-repo-spelling-splits.md
@@ -0,0 +1,180 @@
+- issue: https://github.com/PyAutoLabs/PyAutoBrain/issues/287 (closed on ship)
+- shipped: 2026-08-26 — PyAutoBrain PR https://github.com/PyAutoLabs/PyAutoBrain/pull/288
+ and PyAutoMind PR https://github.com/PyAutoLabs/PyAutoMind/pull/333 (Mind merged first)
+- classification: bug (PyAutoBrain + PyAutoMind; organism infrastructure — neither
+ library nor workspace)
+- summary: The bare and prefixed spellings of seven repos normalised to two different
+ keys, so a policy map filed under one was invisible to the other. The keying question
+ the prompt demanded be answered first was answered — **organs key PREFIXED** — and
+ then the class was closed rather than its fourth instance: the bare/prefixed/package
+ alias join now DERIVES from the body map's new `package:` field, and five guards make
+ the next gap fail loudly.
+
+## The decision, taken before the edit
+
+Organs key on the prefixed form (`pyautobrain`). Not a new rule — #269's rule made
+executable: *the canonical key is the package the repo SHIPS where it ships one, the
+repo name where it does not.* Organs ship no package; Nerves is the one that does and
+already keyed bare (`autonerves`). Everything downstream was already filed prefixed for
+organs — `test_witness`, `target_signals`, `REPO_DISPLAY`, `target_default_wiki`, and
+the Mind's own `draft/*/pyautobrain/` target folders — so keying bare would have rekeyed
+five maps for no gain.
+
+## Shipped changes
+
+- `PyAutoMind/repos.yaml`: `package:` on the seven repos that ship one (six libraries +
+ Nerves), documented in the header as identity — the other name a repo is known by.
+- `PyAutoBrain/agents/faculties/sizing/_sizing.py`: `_body_map_specs`, `canonical_key`,
+ `spellings_of`, `unreachable_repos`, `_derived_aliases`, `_repo_aliases`; the alias
+ table is now derived ∪ hand, with a conflict between them raising rather than one
+ quietly winning.
+- `PyAutoBrain/config/policy.yaml`: `repo_aliases` trimmed to what a body map cannot
+ know, with the canonical-key rule written where the maps live.
+ `extra_organism_targets` emptied — `autohands` is derived now.
+- `PyAutoBrain/agents/conductors/intake/_intake.py`: `REPO_DISPLAY` derived too.
+- `PyAutoBrain/tests/test_policy_seams.py`: five guards (below).
+
+## What the reproduction found beyond the prompt
+
+The prompt named five organs. Sweeping every body-map repo found seven splits — the five
+organs, PyAutoHands (joined only by a policy literal), and **PyAutoScientist** — plus one
+repo that could never resolve at all.
+
+**`pyautolabs.github.io` was DE-REGISTERED, not joined.** This is a deliberate deviation
+from the scope answered at plan time ("join both"). `normalise_repo` truncates at the
+first `.`/`/`, so no mention can reach a dotted repo name. The obvious fix — alias the
+truncated head — is *worse* than the gap: that head is `pyautolabs`, the ORG name, so
+`@PyAutoLabs/PyAutoFit` would have started resolving to the static site instead of
+PyAutoFit. The acceptance criterion's other branch ("or is deliberately not registered")
+covers it. The exclusion derives from the names themselves (any name carrying a
+separator), not a hand-kept list, and a guard pins both halves: the repo is out of the
+known targets AND its truncated head still resolves to nothing.
+
+## Guards, each mutation-tested
+
+| Guard | Mutation that proves it |
+|---|---|
+| no repo splits across two keys | removing the derived join names all seven affected repos |
+| no alias points at a key nothing is filed under | — |
+| body map `package:` agrees with the witness map | mis-keying one `package:` value raises at import |
+| unreachable repos excluded, and their head resolves to nothing | forcing the dotted repo back into the set |
+| canonical keys survive a body map with no `package:` | deleting the pre-package fallback rows |
+
+## Traps and findings
+
+- **The cross-repo CI pinning is the trap this task turns on.** `tests.yml` checks the
+ sibling Mind out at `main`, pinned, so a Brain half that REQUIRED the Mind half would
+ sit red until Mind merged. `canonical_key` falls back to the hand table's library rows
+ when no `package:` is declared, giving identical keys either way — and a test pins that
+ property so the fallback is not an unexercised path. The corroboration guard did NOT
+ have that tolerance on first push and **failed CI for exactly this reason**; the fix
+ made it all-or-nothing (a map declaring no package has nothing to corroborate; one
+ declaring *some* must declare all), which is a stronger guard than the one it replaced,
+ not a weakened one.
+- **The tenant firewall rejected two drafts of the prose**, for naming PyAutoCTI,
+ PyAutoReduce, PyAutoScientist and PyAutoLabs in comments — the third consecutive task
+ where it caught the agent's own drift (cf. #267, #269). The comments read generically
+ because of it.
+- **Behaviour preservation was measured, not asserted.** An A/B over every spelling in
+ the alias table and target sets shows exactly six resolutions changed, all intended
+ joins; libraries, short forms, renames, workspaces and org-qualified paths byte-identical.
+- **`REPO_DISPLAY` had the beginnings of the same drift** — reachable keys (`pyautohands`,
+ the CTI/Reduce libraries, the HowTos) carried no row, so a header rendered
+ `Target: pyautohands`. Newly-reachable keys would have inherited it, so it was derived
+ in the same pass rather than special-cased.
+- **A `git checkout --` during the mutation sweep clobbered the implementation** and it
+ had to be reapplied. Worth naming: mutation-testing a file you are also editing needs a
+ file backup, not a git restore.
+
+## Gate
+
+Brain tests 515 pass (510 baseline + 5 guards), on both matrix legs in CI, and locally
+against BOTH body-map states — the branch's Mind with all seven packages declared, and a
+Mind checkout at `origin/main` with none. `repos_sync --check` all 13 legs OK. Smoke n/a
+(organism repos). **Heart NOT EVALUATED** — `pyauto-heart` is unreachable from a
+web-github session, so leg 4 of the ship gate never ran; recorded that way rather than
+claimed clean. Effective autonomy `supervised`; the run took its plan checkpoint (two
+scope questions put to the human) and proceeded on the answers.
+
+## Follow-ups
+
+- The seven pre-package library rows in `repo_aliases` are the fallback for a body map
+ without `package:`. Now that PyAutoMind#333 has merged they are dead weight and can be
+ deleted, which also fully arms the corroboration guard's strict path.
+- `normalise_repo` still passes `maxsplit` positionally (a DeprecationWarning on 3.13).
+ The copy this task introduced was fixed; the pre-existing one was left alone.
+
+## Original prompt
+
+# Organ repo spellings split across two normalised keys
+
+Type: bug
+Target: pyautobrain
+Repos:
+- PyAutoBrain
+Difficulty: large
+Autonomy: supervised
+Priority: normal
+Status: formalised
+Filed: 2026-08-24
+Issued: 2026-08-26
+
+Found while shipping the refactor witness-map audit (@PyAutoBrain#269 / PR #271).
+
+## The defect
+
+The bare spellings of the organ repos do not normalise to the same key as their
+repo spellings, so a policy map keyed on one is invisible to the other.
+
+@PyAutoBrain/agents/faculties/sizing/_sizing.py (`_target_sets.names_for`)
+registers BOTH spellings of every `PyAuto*` repo as known targets
+(`name.lower()` and `name.lower()[2:]`), but the `repo_aliases` table in
+@PyAutoBrain/config/policy.yaml only joins them for the libraries:
+
+| mention | normalises to | resolves? |
+|---|---|---|
+| `@PyAutoBrain` | `pyautobrain` | yes |
+| `@autobrain` | `autobrain` | no — a known target with nothing filed under it |
+
+Same for the bare Heart, Memory, Mind and Gut spellings. `@autohands` was joined
+in PR #271 because `extra_organism_targets` declares it explicitly and it
+reaches real code today; the rest are latent — no prompt in the Mind history
+uses them yet.
+
+## Decide before fixing
+
+This is NOT a blind alias sweep. Should organs key on the bare form
+(`autobrain`) or the prefixed form (`pyautobrain`)? `target_signals` keys the
+prefixed form (`pyautobrain`, `pyautomind`, `pyautoheart`, `pyautobuild`), and
+after #271 `test_witness` keys organs prefixed but libraries bare. Whichever way
+it goes has to hold across both maps at once, so answer the question first and
+let the edit follow.
+
+## The wider question this raises
+
+Three separate repos have now hit this same defect class — PyAutoNerves
+(#267), PyAutoCTI and PyAutoReduce (#269). The shared cause is that
+`repo_aliases` is HAND-MAINTAINED while the known-target set is DERIVED from the
+body map, so the two drift silently and the gap only ever surfaces as a
+wrong-but-plausible conductor message ("strengthen tests first" for a
+well-tested repo).
+
+Guard 2 added in #271 — a witness-map key must be what its repo normalises to —
+closes the witness-map instance only. An alias gap in a map with no coverage
+guard is still invisible. Consider deriving `repo_aliases` from the body map so
+the class closes rather than its fourth instance.
+
+## Notes on this prompt's header
+
+`pyauto-brain intake classify` scored this `bug` at high confidence (kept), but
+proposed `Target: autocti` — it read the prior-instance history above as the
+subject — and `Difficulty: too-large` (score 10), inflated by the repo names and
+the design-decision keywords. Corrected to `pyautobrain` / `large` on review.
+An alternative classification is `refactor`: the fix is behaviour-preserving for
+every spelling that resolves today. `bug` was kept because the Bug Agent's
+investigate-first strategy matches the "decide before fixing" requirement above.
+
+Acceptance: every spelling of every body-map repo that `_target_sets` registers
+as a known target resolves to a key the policy maps are actually filed under, or
+is deliberately not registered; a guard makes a future alias gap fail loudly
+rather than degrade a conductor's advice.
diff --git a/complete/index.md b/complete/index.md
index bfad2dbe..84c66e86 100644
--- a/complete/index.md
+++ b/complete/index.md
@@ -6,7 +6,7 @@ Token-light navigation over the finished-work records (schema:
only then grep a dated bucket. Curators: edit the band between the CURATED
markers; everything below GENERATED is rebuilt.
-1141 records across 7 buckets.
+1142 records across 7 buckets.
## Highlights
@@ -159,6 +159,7 @@ _(curate hard-won records here — survives regeneration.)_
- [numerical-inversion-failures](2026/08/numerical-inversion-failures.md)
- [one-tap-dashboard-rollout](2026/08/one-tap-dashboard-rollout.md) — the pattern (a generated page + one-tap 📋 copy-for-Claude payloads, phone
- [optional-dependency-skip-guards](2026/08/optional-dependency-skip-guards.md) — `test_nautilus.py::test__single_core_builds_no_pool` had no skip guard
+- [organ-repo-spelling-splits](2026/08/organ-repo-spelling-splits.md) — closed on ship
- [organism-board-final-readmes](2026/08/organism-board-final-readmes.md) — close on record
- [paper-management-pipeline](2026/08/paper-management-pipeline.md) — auto-closed on merge
- [per-parameter-step-scaling](2026/08/per-parameter-step-scaling.md)
diff --git a/dashboard.html b/dashboard.html
index f3e997ba..f1fa40b3 100644
--- a/dashboard.html
+++ b/dashboard.html
@@ -185,7 +185,7 @@
PyAutoMindDashboard
Intent. Priority. Flow.
Every task the Mind is holding. Tap a task's 📋 and its /start_dev command is on your clipboard — paste it into a Claude Code chat to route Claude straight to that task. Recent is the same work by date — what has been happening rather than what to do next.
-
2In flight
3Parked
5Planned
135Backlog
+
1In flight
3Parked
5Planned
135Backlog
Last updated 2026-08-26. This page is generated from active/, draft/ and the registry files, so it is only as current as they are. dashboard_refresh.yml re-renders it on every push to main — that heals a stale page, but not a stale prompt: a task that shipped without its prompt advancing to complete/ keeps rendering here as pickable backlog. Reconciling those is the refresh below.