7907c8d286
Two independent defects, both making the script report a correct checkout as wrong. Neither is new; both were invisible because nobody ran it here. 1. `export/` is the exporter's OUTPUT and is gitignored (`/export*/`). A checkout where nobody has run the exporter has no `export/` at all, so the four `DECISIONS.md`/`BLOCKED.md` citations of `export/manifest.json` and `export/screens/...` landed in "resolve NOWHERE" and the check exited 1 -- red for a state no edit can fix, which is the exact shape its own docstring says it exists to avoid. `check-capture-citations` learned this for `docs/re/captures/`; same rule now: absent BECAUSE UNBUILT is reported, absent while the tree IS built still fails. Verified both ways -- `mkdir export` and the same four go back to failing. 2. The selftest's peer-branch fixture cited `docs/re/f5-a-press-snaps-the-plate.md`, which the consolidation made an ordinary local file. The fixture stopped testing the scanner and started reporting it broken; `PEER_REFS` also still named `origin/auto/frame-blend-draw-path`, a branch that no longer exists. The selftest now FINDS a peer-only path at runtime, and where none exists -- the normal case on a clean checkout, measured: zero -- it says the class is empty here rather than claiming a failure. The class itself stays: the next topic branch that lands a finding recreates the condition exactly. Measured: check 123 citations, 119 resolve, 4 unbuilt, 0 nowhere, exit 0 (was exit 1). Selftest ok (was 🔴 BROKEN, on `main` too). The new generated-tree case was confirmed to FAIL against the unfixed function first. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Sylpheed
A clean-room reverse engineering and port project for Project Sylpheed: Arc of Deception (Xbox 360, 2007).
Three things live here, in one repository so that a change spanning them lands as one commit:
| The decoders | crates/sylpheed-formats — the disc's formats, read and verified disc-wide |
| The port | port/ — a Godot 4 project, plus crates/sylpheed-export which converts a disc into the open asset tree it reads |
| The corpus | docs/re/ — what has been reverse engineered, with its evidence, its retractions and its dead ends |
You need your own copy of the game. No game content is in this repository and none ever will be. The exporter reads the disc you supply.
The oracle is the real game
sylpheed-cliand the Explorer are tools for verifying our decoding. They are hypotheses under test and they have been wrong. When something must be checked against the truth, the truth is the game running in Xenia Canary, captured — not any renderer of ours.
This is stated first because getting it backwards is the most expensive mistake this project has made.
Layout
crates/
sylpheed-formats/ the decoders. Disc-wide verified; the corpus is its spec
sylpheed-cli/ headless tools -- render a screen, dump a table, probe audio
sylpheed-viewer/ the Explorer: a human's window onto the disc. STATIC data only
sylpheed-export/ disc -> the open, moddable asset tree
port/ the Godot 4 project. Reads open formats ONLY
authored/ decisions that are NOT on the disc, each with its reason
data/
base/ generated by the exporter. Gitignored, never hand-edited
mods/ drop-in overrides. Yours
docs/
re/ the corpus: findings, refutations, method traps
game/ how the game is navigated -- menus, modals, flight
port/ the port's mission, its handoff contract, modding rules
-- and RUNNING.md, which is how you actually start it
agents/ how the agent team works together
tools/ capture harnesses, probes, the share tool
exchange/ transient inter-agent files. NOT in git
docker/ the agent containers
Where to start
docs/re/INDEX.md— what is decodeddocs/re/REFUTED.md— what has been tested and dieddocs/re/METHOD.md— traps this project has already paid fordocs/game/navigation.md— how the game is navigateddocs/port/MODDING.md— why the asset tree looks the way it does
Xenia Canary is a separate repository: it is a fork tracking upstream, and it carries our instrumentation.
Conventions
Confidence is per claim, never per document: ✅ CONFIRMED · 🟡 PROBABLE ·
❔ HYPOTHESIS · ❌ REFUTED. A withdrawn result is kept with its reasoning
rather than deleted — that is why the numbers here can be trusted.