erik/acdream
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

docs(p2): session wrap — P1 done, P2 (Path 5 step-up) localized; handoff + plan/CLAUDE.md update

Browse source 
P1 membership is DONE (proven to already match retail; the 0/11 was a cdb capture artifact;
merged + pushed). P2 root cause localized to BSP Path 5 grounded step-up: the Path 5 wrappers
(DoStepUp=retail step_up, DoStepDown=retail step_down) are verified faithful + reached; the
divergence is in the step-up CLIMB (find_walkable/step_sphere_down up-adjust when sp.StepUp=true).

- docs/research/2026-06-03-p2-door-stepup-handoff.md: canonical P2 pickup + fresh-session prompt +
  DO-NOT-RETRY (the wrappers) + the tooling note (xunit swallows Console.WriteLine).
- master-plan §3: P1 marked DONE + the (a)-(d) deletes/unifications re-scoped to approval-gated
  refactors of working code; P2 localization recorded.
- CLAUDE.md M1.5: dated 2026-06-03 pointer (P1 done, P2 active, render seam in P3/P4, pickup doc).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
Erik 2026-06-04 08:52:56 +02:00
parent f0d37d8955
commit 82045805fd
3 changed files with 220 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

24
CLAUDE.md
View file

@ -761,7 +761,29 @@ Apparatus `ACDREAM_PROBE_FLAP` + `tools/A8CellAudit` are committed + ready. Do N
H1 (PVS grounding) or H2 (`PortalSide` side-test) — both evidence-disproven.
**Currently working toward: M1.5 — Indoor world feels right** (resumed
from 2026-05-20 baseline after Phase O ship). **A6.P1 + A6.P2 + A6.P3
from 2026-05-20 baseline after Phase O ship).
**2026-06-03 — P1 membership DONE + P2 active (read this first).** The verbatim spatial-pipeline
port (master plan
[`docs/superpowers/specs/2026-06-03-verbatim-spatial-pipeline-port-master-plan.md`](docs/superpowers/specs/2026-06-03-verbatim-spatial-pipeline-port-master-plan.md))
is the active effort. **P1 (membership) = DONE** — proven to ALREADY match retail; the believed
doorway "0/11 lag" was a cdb CAPTURE ARTIFACT (`change_cell` logs `m_position` before `set_frame`
writes it). Aligned re-capture → production gate 9/9 with NO code change; live `[cell-transit]` clean.
Merged + pushed to both remotes (HEAD `f0d37d8`). **P2 (door/building-shell collision) = IN PROGRESS,
root cause LOCALIZED** to **BSP Path 5 grounded step-up** — the Path 5 wrappers (`DoStepUp`=retail
`step_up`, `DoStepDown`=retail `step_down`) are verified faithful; the divergence is in the step-up
CLIMB (`find_walkable`/`step_sphere_down` up-adjust). The 5 failing Core tests are P2's target. The
visible doorway seam has moved from physics into the RENDER path (P3 camera-collision + P4 PView seal).
The (a)–(d) membership cleanups (remove `ResolveCellId`, unify `find_env_collisions`, intrinsic
building stabs, per-cell collision graph) are **approval-gated refactors of WORKING code**. **CANONICAL
PICKUP:**
[`docs/research/2026-06-03-p2-door-stepup-handoff.md`](docs/research/2026-06-03-p2-door-stepup-handoff.md)
(+ the P1 RESOLVED banner in
[`docs/research/2026-06-03-p1-membership-swept-advance-handoff.md`](docs/research/2026-06-03-p1-membership-swept-advance-handoff.md)
and the visual-gate/render-residuals note
[`docs/research/2026-06-03-p1-visual-gate-render-residuals.md`](docs/research/2026-06-03-p1-visual-gate-render-residuals.md)).
**A6.P1 + A6.P2 + A6.P3
slice 1 SHIPPED 2026-05-21.** **A6.P3 slice 2 v2 SHIPPED 2026-05-22**
(commit `f8d669b`): tried removing the L622 per-tick CP seed
(`892019b` v1) but it broke BSP step_up at the last step of stairs;