diff --git a/.gitattributes b/.gitattributes
index 290ad9d0..a546dd49 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -1,6 +1,8 @@
 * text=auto eol=lf
 *.png binary
 *.jpg binary
+*.gif binary
+*.mp4 binary
 *.z64 binary
 *.exe binary
 *.bin binary
diff --git a/.github/ISSUE_TEMPLATE/bug_report.yml b/.github/ISSUE_TEMPLATE/bug_report.yml
new file mode 100644
index 00000000..1c0d8950
--- /dev/null
+++ b/.github/ISSUE_TEMPLATE/bug_report.yml
@@ -0,0 +1,71 @@
+name: Bug report (build or run)
+description: A build failure, crash, hang, or rendering/audio/input defect in the PC port
+labels: ["bug"]
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Thanks for filing a report. First, please check:
+
+        - The port is **Phase 2 of 4** — rendering is incomplete and many
+          cosmetic defects are already tracked in
+          [`docs/dev/GRAPHICS-BACKLOG.md`](https://github.com/jkdansereau/goldeneye-pc-port/blob/main/docs/dev/GRAPHICS-BACKLOG.md).
+          Search open issues and that file before filing.
+        - Only the **Windows / MSYS2 MINGW64** build is tested. Linux and
+          macOS are best-effort — build breaks there are expected.
+        - Do **not** attach a ROM, extracted assets, or any game data.
+  - type: input
+    id: commit
+    attributes:
+      label: Commit
+      description: Output of `git rev-parse --short HEAD`
+      placeholder: 311fd012
+    validations:
+      required: true
+  - type: dropdown
+    id: region
+    attributes:
+      label: Region / ROMID
+      options: [ntsc-final, pal-final, jpn-final]
+    validations:
+      required: true
+  - type: input
+    id: os
+    attributes:
+      label: OS / environment
+      description: e.g. "Windows 10, MSYS2 MINGW64" or "Ubuntu 22.04, native gcc"
+    validations:
+      required: true
+  - type: dropdown
+    id: stage
+    attributes:
+      label: Which stage fails?
+      options:
+        - Asset extraction
+        - CMake configure
+        - Compile / link
+        - Runtime — crash or hang
+        - Runtime — wrong rendering / audio / input
+    validations:
+      required: true
+  - type: textarea
+    id: what
+    attributes:
+      label: What happened vs. what you expected
+    validations:
+      required: true
+  - type: textarea
+    id: repro
+    attributes:
+      label: Steps to reproduce
+      description: Exact commands, flags (e.g. `-level_09`), and what you did in-game.
+    validations:
+      required: true
+  - type: textarea
+    id: logs
+    attributes:
+      label: Logs
+      description: |
+        Paste the terminal error, and `ge007.crash.log` if the port crashed.
+        For a rendering defect, attach a screenshot.
+      render: text
diff --git a/.github/ISSUE_TEMPLATE/config.yml b/.github/ISSUE_TEMPLATE/config.yml
new file mode 100644
index 00000000..2edadabd
--- /dev/null
+++ b/.github/ISSUE_TEMPLATE/config.yml
@@ -0,0 +1,14 @@
+blank_issues_enabled: true
+contact_links:
+  - name: Build & setup guide
+    url: https://github.com/jkdansereau/goldeneye-pc-port/blob/main/docs/building.md
+    about: Read this first for build/extraction problems — most are covered here.
+  - name: Known rendering defects
+    url: https://github.com/jkdansereau/goldeneye-pc-port/blob/main/docs/dev/GRAPHICS-BACKLOG.md
+    about: Cosmetic / rendering issues already known and tracked. Check before filing.
+  - name: Security vulnerability
+    url: https://github.com/jkdansereau/goldeneye-pc-port/security/advisories/new
+    about: Report memory-safety / build-pipeline vulnerabilities privately, not as a public issue.
+  - name: Rights-holder concern
+    url: https://github.com/jkdansereau/goldeneye-pc-port/blob/main/NOTICE
+    about: If you represent a rights holder, see NOTICE and open an issue titled accordingly.
diff --git a/.github/SECURITY.md b/.github/SECURITY.md
new file mode 100644
index 00000000..5901c13f
--- /dev/null
+++ b/.github/SECURITY.md
@@ -0,0 +1,41 @@
+# Security Policy
+
+This project is a non-commercial, fan-made research port. It has no server
+component and ships no binaries. The realistic security surface is:
+
+- the PC port executable parsing your own ROM and asset files at load time
+  (a malformed ROM/asset could in principle crash it or worse), and
+- the build/extraction scripts and CI workflow.
+
+## Reporting a vulnerability
+
+Please **do not** open a public issue for a security problem.
+
+Use GitHub's private vulnerability reporting:
+**Security → Report a vulnerability** on this repository
+(<https://github.com/jkdansereau/goldeneye-pc-port/security/advisories/new>).
+
+If that is unavailable, email the maintainer at the address on their GitHub
+profile with `SECURITY` in the subject.
+
+Please include:
+
+- affected version / commit hash (from `git rev-parse HEAD`),
+- OS and how you built (region, `IDO_RECOMP`, MSYS2 vs WSL, …),
+- a minimal reproduction, and
+- the crash log (`ge007.crash.log`) or a stack trace if you have one.
+
+## Scope
+
+In scope: memory-safety bugs in the `port/` layer and PC-port tooling,
+issues in the CI workflow or build scripts, and dependency problems we can
+act on.
+
+Out of scope: bugs inherited unchanged from the upstream
+[GoldenEye 007 decompilation](https://github.com/n64decomp/007) that are not
+made worse by the port (report those upstream), missing-asset or wrong-ROM
+errors, and anything requiring a ROM or assets we do not distribute.
+
+## Supported versions
+
+Only the tip of the default branch is supported. There are no releases yet.
diff --git a/.github/dependabot.yml b/.github/dependabot.yml
new file mode 100644
index 00000000..e5c49c41
--- /dev/null
+++ b/.github/dependabot.yml
@@ -0,0 +1,8 @@
+version: 2
+updates:
+  # Keep the GitHub Actions used by .github/workflows/ up to date. Opens a PR
+  # when a new version of an action is released; review and merge like any other.
+  - package-ecosystem: github-actions
+    directory: /
+    schedule:
+      interval: monthly
diff --git a/.github/labels.yml b/.github/labels.yml
new file mode 100644
index 00000000..7f260a01
--- /dev/null
+++ b/.github/labels.yml
@@ -0,0 +1,58 @@
+# Source of truth for this repo's issue/PR labels.
+# GitHub does not read this file automatically. Apply it with the gh commands
+# in the comment block at the bottom, or wire up a label-sync action later.
+#
+# `bug` and `documentation` are GitHub defaults and are kept as-is.
+
+- name: bug
+  color: d73a4a
+  description: Something is broken
+
+- name: rendering
+  color: 5319e7
+  description: Graphics / fast3d / RSP output defect (see docs/dev/GRAPHICS-BACKLOG.md)
+
+- name: audio
+  color: 1d76db
+  description: Sound / music / libaudio (Phase 3)
+
+- name: input
+  color: 0e8a16
+  description: Keyboard / mouse / controller
+
+- name: build
+  color: fbca04
+  description: CMake / MSYS2 / toolchain / CI
+
+- name: asset-extraction
+  color: c2e0c6
+  description: ROM asset extraction pipeline (scripts/, tools/extractor)
+
+- name: upstream
+  color: bfd4f2
+  description: Originates in the n64decomp/007 decompilation, not the port — may belong upstream
+
+- name: abi-64bit
+  color: fef2c0
+  description: 32-to-64-bit pointer-width / struct-layout issue (the #ifdef PORT exception)
+
+- name: needs-repro
+  color: e99695
+  description: Cannot reproduce yet; waiting on more detail
+
+- name: wontfix
+  color: ffffff
+  description: Out of scope or working as intended
+
+# Apply with the GitHub CLI (run once, after the repo exists):
+#
+#   gh label create rendering       -c 5319e7 -d "Graphics / fast3d / RSP output defect"
+#   gh label create audio           -c 1d76db -d "Sound / music / libaudio (Phase 3)"
+#   gh label create input           -c 0e8a16 -d "Keyboard / mouse / controller"
+#   gh label create build           -c fbca04 -d "CMake / MSYS2 / toolchain / CI"
+#   gh label create asset-extraction -c c2e0c6 -d "ROM asset extraction pipeline"
+#   gh label create upstream        -c bfd4f2 -d "Originates in the decompilation, not the port"
+#   gh label create abi-64bit       -c fef2c0 -d "32-to-64-bit pointer-width / struct-layout issue"
+#   gh label create needs-repro     -c e99695 -d "Cannot reproduce yet; waiting on more detail"
+#
+# `bug` and `wontfix` already exist as GitHub defaults.
diff --git a/.github/pull_request_template.md b/.github/pull_request_template.md
new file mode 100644
index 00000000..193c3394
--- /dev/null
+++ b/.github/pull_request_template.md
@@ -0,0 +1,37 @@
+<!--
+Read CONTRIBUTING.md first. This is a faithfulness-focused port; the ground
+rules there are non-negotiable.
+-->
+
+## What this changes
+
+<!-- One or two sentences. Link the issue it closes: "Closes #123". -->
+
+## Why
+
+<!-- The reasoning, not just the diff. What was wrong / missing? -->
+
+## Scope check
+
+- [ ] No changes under `src/` or `include/` — **or** the only changes are the
+      narrow `#ifdef PORT` ABI exception (CONTRIBUTING.md rule 2), and each is
+      documented in `docs/porting-notes.md` / `docs/dev/findings.md`.
+- [ ] `Makefile`, `tools/`, `rsp/`, `ld/` untouched (N64 build).
+- [ ] If `CMakeLists.txt` `REGION_DEFS` changed, it still matches the N64
+      `Makefile` per-region macro set exactly.
+
+## Verification
+
+<!-- What you actually ran. Delete lines that don't apply. -->
+
+- [ ] `./build-pc.sh ntsc-final` — clean configure + link
+- [ ] Crash-free run of at least one level (`-level_09`)
+- [ ] Single-frame `GE_PCDUMP` diff against the committed golden — no
+      unexpected change
+- [ ] pal-final / jpn-final also configured
+
+Platform tested: <!-- e.g. Windows 10 / MSYS2 MINGW64 -->
+
+## Notes for the reviewer
+
+<!-- Anything uncertain, follow-ups, or areas that need a closer look. -->
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index f73047bc..ffa2c51b 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -2,26 +2,43 @@ name: CI
 
 on:
   push:
-    branches: [ master ]
+    branches: [ main, master ]
     paths-ignore: [ '**.md', 'docs/**', 'LICENSE', 'NOTICE' ]
   pull_request:
+    paths-ignore: [ '**.md', 'docs/**', 'LICENSE', 'NOTICE' ]
   workflow_dispatch:
 
-# The full port build needs assets extracted from a copyrighted ROM, which CI
-# does not have. This job validates everything that does NOT depend on those
-# assets: the PC-port tooling, the generated ROM-asset symbol files, and a
-# CMake configure of all three regions (which checks REGION_DEFS, the source
-# list, and that SDL2 / zlib / OpenGL are discoverable).
+# Minimal token: CI only reads the repo and uploads build artifacts, neither of
+# which needs write access. Fork PRs run untrusted build scripts on the runner,
+# so keep the granted scope as small as possible and never expose secrets here.
+permissions:
+  contents: read
+
+concurrency:
+  group: ci-${{ github.ref }}
+  cancel-in-progress: true
+
+# Two jobs:
 #
-# This is a configure-only smoke test. It does NOT compile or run the port, and
-# it running on ubuntu is NOT a statement that the port builds or works on
-# Linux — it has only ever been built and tested on Windows (MSYS2).
+#   validate      — ubuntu, configure-only. Byte-compiles the tooling,
+#                   regenerates the US ROM-asset symbol file, and CMake-
+#                   configures ntsc-final (REGION_DEFS, the source list, and
+#                   that SDL2 / zlib / OpenGL are discoverable). No ROM is used.
+#                   Does NOT compile; running on ubuntu is NOT a claim the port
+#                   works on Linux.
+#
+#   windows-build — windows/MSYS2, the real supported platform. Compiles and
+#                   links the full executable. Neither the ROM nor extracted
+#                   assets are needed to BUILD (they are runtime-only), so CI
+#                   builds but never runs the port.
 
 jobs:
   validate:
     runs-on: ubuntu-22.04
     steps:
       - uses: actions/checkout@v4
+        with:
+          persist-credentials: false
 
       - name: Install dependencies
         run: |
@@ -31,17 +48,61 @@ jobs:
       - name: Byte-compile Python tooling
         run: python3 -m compileall -q scripts tools_pc
 
-      - name: Regenerate ROM-asset symbol files
+      # US only. gen_romassets.py resolves every symbol for the US region from
+      # the committed CSVs with no ROM present; PAL/JP each need a handful of
+      # region-specific prop/font offsets that only come from that region's ROM
+      # (see docs/building.md — PAL/JP builders generate romassets_e/j.s
+      # locally). No ROM is used or required here.
+      - name: Regenerate ROM-asset symbol file (US)
         run: |
           python3 scripts/gen_romassets.py u
-          python3 scripts/gen_romassets.py e
-          python3 scripts/gen_romassets.py j
           git diff --stat -- port/src/romassets_u.s
 
-      - name: CMake configure (all regions)
-        run: |
-          for r in ntsc-final pal-final jpn-final; do
-            echo "::group::configure $r"
-            cmake -S . -B "build-$r" -DROMID="$r"
-            echo "::endgroup::"
-          done
+      - name: CMake configure (ntsc-final)
+        run: cmake -S . -B build-ntsc -DROMID=ntsc-final
+
+  # Real Windows (MSYS2 MINGW64) build — the only platform the port is
+  # developed and tested on. This compiles and LINKS the full executable.
+  #
+  # No ROM and no extracted assets are needed to build: the compiled asset set
+  # is a handful of committed pure-data .c files plus the generated
+  # address-constant romassets_<region>.s. The ROM (.z64) and the ROM-derived
+  # blobs are runtime-only dependencies — CI cannot run the game, only build it.
+  windows-build:
+    runs-on: windows-latest
+    defaults:
+      run:
+        shell: msys2 {0}
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          persist-credentials: false
+
+      - uses: msys2/setup-msys2@v2
+        with:
+          msystem: MINGW64
+          update: true
+          install: >-
+            mingw-w64-x86_64-toolchain
+            mingw-w64-x86_64-cmake
+            mingw-w64-x86_64-SDL2
+            mingw-w64-x86_64-zlib
+            mingw-w64-x86_64-python
+
+      - name: Regenerate ROM-asset symbol file (US)
+        run: python scripts/gen_romassets.py u
+
+      - name: Build (ntsc-final)
+        run: ./build-pc.sh ntsc-final
+
+      - name: Upload executable
+        uses: actions/upload-artifact@v4
+        with:
+          name: ge007-ntsc-final
+          path: build-pc/ge007*
+          if-no-files-found: error
+          retention-days: 14
+
+      # CI builds the executable but does not run it — running the port needs a
+      # ROM, which CI does not provide. Runtime checks (the -level_09 soak, a
+      # golden-frame diff) would run on a self-hosted runner instead. Not planned.
diff --git a/.gitignore b/.gitignore
index 30fe2601..eb65ba94 100644
--- a/.gitignore
+++ b/.gitignore
@@ -76,6 +76,12 @@ ge007.ini
 !build/test.c
 !notes/StyleGuide.md
 #tools
+# Original SGI IRIX 5.3 IDO toolchain (proprietary SGI binaries) and the SGI
+# image viewer are NOT distributed with this repo. They are only needed to
+# build the N64 ROM or to regenerate tools/ido5.3_recomp; the PC port does not
+# use them. See docs/SetupGuide.md "Recompile IDO" for how to obtain them.
+tools/irix/
+tools/SGIImageViewer.exe
 tools/extractor/*.o
 tools/extractor/extractor
 tools/aaa_rip/main.o
@@ -117,7 +123,6 @@ assets/obseg/em/data/under.seg
 
 scripts/local/
 scripts/make/local/
-scripts/__pycache__/generate_prop_model_c.cpython-310.pyc
 
 # PC port debug/session artifacts
 _run.log
diff --git a/AGENTS.md b/AGENTS.md
index 76fa8ee9..9729eb3d 100644
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -4,8 +4,8 @@
 
 - `n64decomp/007`: WIP decompilation of GoldenEye 007 (N64), byte-matches US/EU/JP ROMs.
 - Active work: **PC port** modelled on the Perfect Dark PC port (same Rare "Indy" engine family).
-- **Reference docs:** `docs/internals.md` — architecture, GE-specific RSP deltas, phased plan (§1–§10). `docs/dev/findings.md` — the `Dxx` finding log (§F + §H, indexed at the top of §F); read the entry you need, don't linear-read. `docs/porting-notes.md` — the recurring N64→PC bug classes, one screen.
-- **Current status:** the README "Status" section. Current task + environment: `docs/HANDOFF.md` (a rolling local working file).
+- **Reference docs:** `docs/internals.md` — architecture, GE-specific RSP deltas, phased plan (§1–§10). `docs/dev/findings.md` — the `Dxx` finding log (§F + §H, indexed at the top of §F); read the entry you need, don't linear-read. `docs/porting-notes.md` — the recurring N64→PC bug classes (dense; skim the headers, read what's relevant).
+- **Current status:** the README "Status" section, and `docs/dev/LEVEL-STATUS.md` for the per-level sweep. Current task + environment: `docs/HANDOFF.md` (a rolling local working file — may be absent in a fresh clone; fall back to the README "Status" section).
 - **Dispatching subagents?** `docs/dev-process.md` — task budgets/deadlines, file partitioning, pre-flight, the standard brief template. Every investigation subagent reads `docs/porting-notes.md` first and appends to it.
 
 ## Non-negotiables
@@ -55,7 +55,7 @@ Run `/linkcheck` for this sweep. Record new findings in `docs/dev/findings.md` 
   start) is functional; file-backed EEPROM saves work. Cosmetic defects are
   parked in `docs/dev/GRAPHICS-BACKLOG.md`.
 - **Phase 3 (audio + input):** input layer done (`port/src/input.c`); polish
-  bugs open (D118*, D169). Audio not started — libaudio → SDL, copy-and-adapt
-  from the PD port's `mixer.c`.
+  bugs open (D118* mouse-look residuals; interactive feel-checks owed). Audio
+  not started — libaudio → SDL, copy-and-adapt from the PD port's `mixer.c`.
 - **Phase 4 (saves + polish):** file-backed EEPROM done; widescreen, config,
   rebinding UI outstanding.
diff --git a/CLAUDE.md b/CLAUDE.md
index 078dadb9..6888f20a 100644
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -12,15 +12,16 @@ Context is scarce. Load by tier; do not blind-read whole files.
 
 - **Tier 0 — always (auto-loaded):** `CLAUDE.md`, `AGENTS.md`. Nothing
   else auto-loads.
-- **Tier 1 — every session start, read fully (both are short):**
+- **Tier 1 — every session start:**
   `docs/HANDOFF.md` (current state + next task + environment — a rolling
   local working file; may be absent in a fresh clone, in which case read
-  the README "Status" section instead),
-  `docs/porting-notes.md` (recurring bug classes).
+  the README "Status" section instead) — read fully;
+  `docs/porting-notes.md` (recurring bug classes) — skim the section
+  headers, read the classes relevant to the task.
 - **Tier 2 — on demand only, do NOT read start-to-finish:**
-  - `docs/internals.md` — architecture / RSP deltas / phased plan (§1–§10
-    of the old research doc). Read the section you need.
-  - `docs/dev/findings.md` — the `Dxx` finding log (§11 + §F + §H).
+  - `docs/internals.md` — architecture / RSP deltas / phased plan (§1–§10).
+    Read the section you need.
+  - `docs/dev/findings.md` — the `Dxx` finding log (§F + §H).
     **Jump to a specific `Dxx` via the index at the top of §F.** Never
     linear-read.
   - `docs/dev/HANDOFF-ARCHIVE.md` — prior-session narrative (M-2…). Only
@@ -33,8 +34,9 @@ Context is scarce. Load by tier; do not blind-read whole files.
 Having a concrete "next step" from the handoff is NOT license to skip this.
 
 1. Read the Tier 1 docs: `docs/HANDOFF.md` (immediate task, standing
-   procedure, environment gotchas) and `docs/porting-notes.md`
-   (recurring bug classes — you will re-derive catalogued bugs if you don't).
+   procedure, environment gotchas — or the README "Status" section if it is
+   absent) and `docs/porting-notes.md` (recurring bug classes — you will
+   re-derive catalogued bugs if you don't).
 2. Skim the README "Status" section for where the project stands.
 3. If you may dispatch a subagent this session, read `docs/dev-process.md`
    NOW — not at dispatch time.
diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index 537e76a4..46a76632 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -11,7 +11,7 @@ keep the port faithful to the original game.
    belong to the N64 build and are never touched for PC work. If PC code seems
    to need a behavioural change in the game, the fix belongs in `port/`.
 
-2. **Narrow ABI exception.** The 32->64-bit transition forces a small class of
+2. **Narrow ABI exception.** The 32→64-bit transition forces a small class of
    mechanical, semantics-preserving edits to ROM-serialized structs (storing an
    embedded 32-bit pointer as `u32` and casting at the use site). These are
    allowed, must be guarded with `#ifdef PORT`, must change no behaviour, and
@@ -35,6 +35,6 @@ keep the port faithful to the original game.
 ## Where to look
 
 - `docs/internals.md` — how the port is structured.
-- `docs/porting-notes.md` — the recurring bug classes; check here before
-  debugging a crash, you are likely looking at a known pattern.
+- `docs/porting-notes.md` — the recurring bug classes. Check here before
+  debugging a crash; you are likely looking at a known pattern.
 - `docs/dev/` — the full finding log and per-level status.
diff --git a/README.md b/README.md
index b045262e..e10d0975 100644
--- a/README.md
+++ b/README.md
@@ -1,5 +1,10 @@
 # GoldenEye 007 — PC port
 
+![status](https://img.shields.io/badge/status-Phase_2_of_4_(rendering)-orange)
+![tested on](https://img.shields.io/badge/tested_on-Windows_x86--64-blue)
+![license](https://img.shields.io/badge/license-MIT-green)
+![built with](https://img.shields.io/badge/built_with-coding_agents-8A2BE2)
+
 A work-in-progress native port of **GoldenEye 007** (Rare, 1997) to modern
 desktop platforms, built on top of the
 [GoldenEye decompilation](https://github.com/n64decomp/007).
@@ -11,26 +16,33 @@ sources are compiled for the host; the N64's Reality Signal Processor is
 emulated in software; every other hardware surface (video, audio, input,
 timers, save storage) is shimmed in a dedicated `port/` layer.
 
+> [!IMPORTANT]
 > **You must supply your own GoldenEye 007 ROM.** This repository contains no
 > Nintendo code or assets, and no ROM. Nothing here is distributable as a
 > playable game — see [Requirements](#requirements) and [Legal](#legal).
 
 <p align="center">
-  <img src="docs/img/archives-1.png" width="49%" alt="Archives level rendering in the port">
-  <img src="docs/img/bunker1-1.png" width="49%" alt="Bunker 1 rendering in the port">
-  <br>
-  <img src="docs/img/archives-2.png" width="32%" alt="Guards in the Archives level">
-  <img src="docs/img/bunker1-2.png" width="32%" alt="Bunker 1 with HUD">
-  <img src="docs/img/bunker1-3.png" width="32%" alt="Bunker 1 storage room">
-  <br><em>In-engine captures &mdash; Archives and Bunker&nbsp;1 running in the port.</em>
+  <img src="docs/img/attract-bunker1.png" width="32%" alt="Bunker 1 intro camera">
+  <img src="docs/media/goldeneye-demo.gif" width="32%" alt="~15 s of the port running: mission dossier, Facility, Silo, Jungle, Archives">
+  <img src="docs/img/attract-dam.png" width="32%" alt="Dam intro camera">
+  <br><em>In-engine, running in the port — Bunker&nbsp;1 and Dam attract views, and a
+  ~15&nbsp;s clip (no audio yet, Phase&nbsp;3): mission dossier &rarr; Facility &rarr; Silo &rarr;
+  Jungle &rarr; Archives.</em>
 </p>
 
-> **Not playable yet.** This is a work-in-progress research port. The intro and
-> menus render and all 21 solo levels load, but it is **not** a complete or
-> reliably playable game: there is **no audio**, the front-end 3D models and
-> some text are broken, input has known rough edges, and you should expect
-> crashes and glitches once you are past the level intro. Treat it as a
-> technical demo of the porting work, not a way to play GoldenEye.
+## Contents
+
+- [Background](#background)
+- [Status](#status)
+- [Requirements](#requirements)
+- [Building](#building) · [Windows (MSYS2)](#windows-msys2) · [Linux](#linux)
+- [Running](#running) · [Default controls](#default-controls)
+- [How it works](#how-it-works)
+- [Project layout](#project-layout)
+- [Documentation](#documentation)
+- [Credits](#credits)
+- [Legal](#legal)
+- [License](#license)
 
 ## Background
 
@@ -55,33 +67,42 @@ It used two agents:
 The two **handed work back and forth** through shared written artifacts:
 
 ```mermaid
-flowchart LR
-    H["human<br/>direction · integration<br/>build + playtest"]
+flowchart TD
+    H["human<br/>direction · integration · playtest"]
     C["Claude<br/>(Claude Code)"]
     Q["Qwen 3.8<br/>(pi, local RTX 5090)"]
-    D[("shared artifacts<br/>finding log · bug-class notes · handoff doc")]
+    D[("shared artifacts<br/>findings · notes · handoff doc")]
     H -->|"scoped task + budget"| C
     H -->|"scoped task + budget"| Q
-    C <-->|read / append| D
-    Q <-->|read / append| D
-    C -.->|"hit usage limit ->"| Q
-    Q -.->|"needs deeper reasoning ->"| C
-    C -->|"patch + write-up"| H
-    Q -->|"patch + write-up"| H
+    C -->|"patch + write-up"| D
+    Q -->|"patch + write-up"| D
+    C <-.->|handoff| Q
+    D --> H
 ```
 
 The handoff document was originally a session-to-session note; it became the
 **interface between the two models** — when Claude hit a usage limit
 mid-problem, the local model picked the task up from that state and continued.
 
-**By the numbers:** ~2 weeks, one person part-time · 217 commits · 160
-root-caused bugs logged · ~30 handoff sessions · `#ifdef PORT` ABI edits in 63
-game-source files · ~17k lines of port layer + ~6k lines of Python asset
-tooling. By day 4 the full ~230-unit codebase compiled and linked; by day 8
-the intro rendered; by **day 13 all 21 solo missions ran crash-free**; by
-day 14 the front end was playable end to end. Estimated effort split
-(milestone-weighted): **~60% Claude / ~40% local model** — the local model
-built the entire foundation (build, boot, RSP wiring, converters).
+**By the numbers:**
+
+| Metric | Value |
+|---|---|
+| Timeline | ~2 weeks, one person part-time |
+| Commits | ~223 |
+| Root-caused bugs logged | 162 (`D1`–`D169`; some later merged or withdrawn) |
+| Handoff sessions | ~33 |
+| ABI edits to game code | 63 files, all `#ifdef PORT` |
+| Port layer / tooling | ~17k lines C/C++ · ~6k lines Python |
+| Effort split (milestone-weighted) | ~60% Claude · ~40% local model |
+
+The local model built the entire foundation — build, boot chain, RSP wiring,
+asset converters — then both agents ran the debugging phase together:
+
+- **Day 4** — the full ~230-unit codebase compiles and links
+- **Day 8** — the whole intro renders
+- **Day 13** — all 21 solo missions run crash-free
+- **Day 14** — front end playable end to end
 
 Full write-up, timeline chart, and an honest "what worked / what didn't":
 [`docs/dev/agentic-development.md`](docs/dev/agentic-development.md). The
@@ -89,13 +110,21 @@ workflow itself: [`docs/dev-process.md`](docs/dev-process.md).
 
 ## Status
 
+> [!WARNING]
+> **Not playable yet.** This is a work-in-progress research port. The intro and
+> menus render and all 21 solo levels load, but it is **not** a complete or
+> reliably playable game: there is **no audio**, the front-end 3D models and
+> some text are broken, input has known rough edges, and you should expect
+> crashes and glitches once you are past the level intro. Treat it as a
+> technical demo of the porting work, not a way to play GoldenEye.
+
 **Phase 2 of 4 (rendering).** The port boots, renders, and is playable through
 the front end into the early game. It is not finished and it is not stable.
 
 **Working**
 
-- Boot -> Rare/Nintendo logos -> gun-barrel -> cast intro, fully rendered.
-- Front end: main menu -> mission select -> difficulty -> briefing -> mission start.
+- Boot → Rare/Nintendo logos → gun-barrel → cast intro, fully rendered.
+- Front end: main menu → mission select → difficulty → briefing → mission start.
 - **All 21 solo missions load, render, and survive an unattended play window
   without crashing** (see [`docs/dev/LEVEL-STATUS.md`](docs/dev/LEVEL-STATUS.md)).
 - Software RSP (fast3d): textured world geometry, skeletal characters, HUD,
@@ -107,8 +136,9 @@ the front end into the early game. It is not finished and it is not stable.
 **Not yet working**
 
 - **Audio** — not implemented (Phase 3). The game runs silent.
-- Front-end 3D models (spinning Nintendo logo, gun-barrel Bond, cast) are
-  mispositioned or absent.
+- Some front-end 3D models — the spinning Nintendo logo, and the MISSION
+  COMPLETE / mode-select models — are mispositioned or absent. (The
+  gun-barrel Bond intro renders correctly.)
 - Assorted cosmetic defects (some text layout, a few incomplete asset
   conversions) are tracked in
   [`docs/dev/GRAPHICS-BACKLOG.md`](docs/dev/GRAPHICS-BACKLOG.md) and parked
@@ -173,7 +203,8 @@ cd goldeneye-pc-port
 ./build-pc.sh ntsc-final
 ```
 
-The executable is written to `build-pc/ge007.x86_64` (Windows: `.exe`).
+The executable is written to `build-pc/ge007.x86_64` (on Windows,
+`build-pc/ge007.x86_64.exe`).
 
 ## Running
 
@@ -187,16 +218,16 @@ Configuration is written to `ge007.ini` on first run.
 
 ### Default controls
 
-| Action            | Keyboard / mouse        | Controller            |
-|-------------------|-------------------------|-----------------------|
-| Move / strafe     | `W` `A` `S` `D` / arrows | Left stick            |
-| Aim / look        | Mouse                   | Right stick           |
-| Fire (Z)          | Left mouse / `LCtrl`    | Right trigger         |
-| Aim mode (R)      | Right mouse / `LShift`  | Left trigger          |
-| Use / accept (A)  | `Space` / `E`           | A / X                 |
-| Reload / cancel (B)| `R` / `F`              | B / Y / RB            |
-| L trigger         | `Q`                     | LB                    |
-| Start             | `Enter` / `Tab`         | Start                 |
+| Action              | Keyboard / mouse         | Controller    |
+|---------------------|--------------------------|---------------|
+| Move / strafe       | `W` `A` `S` `D` / arrows  | Left stick    |
+| Aim / look          | Mouse                    | Right stick   |
+| Fire (Z)            | Left mouse / `LCtrl`     | Right trigger |
+| Aim mode (R)        | Right mouse / `LShift`   | Left trigger  |
+| Use / accept (A)    | `Space` / `E` / `X`            | A / X         |
+| Reload / cancel (B) | `R` / `F`                | B / Y / RB    |
+| L trigger           | `Q`                      | LB            |
+| Start               | `Enter` / `Tab`          | Start         |
 
 Mouse sensitivity, Y-inversion and the aim/turn split are tunable in the
 `[Input]` section of `ge007.ini`.
@@ -217,7 +248,7 @@ touch N64 hardware is redirected into `port/`:
 - **`port/src/{video,audio,input,fs,romdata,config}.c`** — the SDL2 / OpenGL /
   filesystem backends.
 
-The 32->64-bit transition forces a small, catalogued class of mechanical
+The 32→64-bit transition forces a small, catalogued class of mechanical
 ABI-only edits to ROM-serialized structs (pointer-width reconciliation); these
 change no behaviour and are documented individually.
 
@@ -244,21 +275,15 @@ docs/               see below
 
 ## Documentation
 
-- [`docs/building.md`](docs/building.md) — full build + asset-extraction guide.
-- [`docs/internals.md`](docs/internals.md) — architecture, the RSP-emulation
-  approach, GE-vs-PD engine differences, the phased plan.
-- [`docs/porting-notes.md`](docs/porting-notes.md) — the recurring N64->PC bug
-  classes hit during the port, with fixes.
-- [`docs/dev/agentic-development.md`](docs/dev/agentic-development.md) — the
-  research angle: the two-agent setup, timeline, handoff workflow, and a
-  candid assessment of what did and didn't work.
-- [`docs/dev-process.md`](docs/dev-process.md) — the investigation workflow in
-  detail (budgets, file partitioning, the finding-log discipline).
-- [`docs/dev/`](docs/dev/) — the raw engineering record: the full finding log,
-  per-level status, graphics backlog, playtest matrices.
-- [`docs/SetupGuide.md`](docs/SetupGuide.md),
-  [`docs/StructureGuide.md`](docs/StructureGuide.md),
-  [`docs/StyleGuide.md`](docs/StyleGuide.md) — inherited from the decompilation.
+| Doc | What's in it |
+|---|---|
+| [`docs/building.md`](docs/building.md) | Full build + asset-extraction guide. |
+| [`docs/internals.md`](docs/internals.md) | Architecture, the RSP-emulation approach, GE-vs-PD engine differences, the phased plan. |
+| [`docs/porting-notes.md`](docs/porting-notes.md) | The recurring N64→PC bug classes hit during the port, with fixes. |
+| [`docs/dev/agentic-development.md`](docs/dev/agentic-development.md) | The research angle: the two-agent setup, timeline, handoff workflow, and an assessment of what did and didn't work. |
+| [`docs/dev-process.md`](docs/dev-process.md) | The investigation workflow in detail — budgets, file partitioning, the finding-log discipline. |
+| [`docs/dev/`](docs/dev/) | The raw engineering record: the full finding log, per-level status, graphics backlog, playtest matrices. |
+| [`docs/SetupGuide.md`](docs/SetupGuide.md), [`docs/StructureGuide.md`](docs/StructureGuide.md), [`docs/StyleGuide.md`](docs/StyleGuide.md) | Inherited from the decompilation. |
 
 ## Credits
 
@@ -267,7 +292,7 @@ This port is a thin layer on a large amount of other people's work.
 **Prior work it is built on**
 
 - The [GoldenEye 007 decompilation](https://github.com/n64decomp/007) — years of
-  effort by kholdfuzion, Larry Ficken, and the project's contributors; plus
+  effort by Larry Ficken ("kholdfuzion") and the project's contributors; plus
   zoinkity's GoldenEye documentation, which the decomp started from. This port
   is a fork of that repository.
 - The [Perfect Dark PC port](https://github.com/fgsfdsfgs/perfect_dark)
@@ -298,8 +323,11 @@ This port is a thin layer on a large amount of other people's work.
 
 **Tools and models used to develop the port**
 
-- [Qwen 3.8](https://github.com/QwenLM/Qwen) (Alibaba Qwen team);
-  [Unsloth](https://unsloth.ai) (the GGUF quantisation and Unsloth Desktop).
+- Qwen 3.8 (Alibaba Qwen team), run locally as the
+  [`unsloth/Qwen3.8-27B-GGUF`](https://huggingface.co/unsloth/Qwen3.8-27B-GGUF)
+  `UD-Q4_K_XL` quant;
+  [Unsloth](https://unsloth.ai) (the GGUF quantisation and Unsloth Desktop,
+  used as the local model server).
 - [pi](https://pi.dev/) — the local coding-agent harness.
 - [Claude / Claude Code](https://claude.com/claude-code) (Anthropic).
 
diff --git a/build-pc.sh b/build-pc.sh
index 9110d42d..8f71030f 100644
--- a/build-pc.sh
+++ b/build-pc.sh
@@ -5,14 +5,14 @@
 # Usage:
 #   ./build-pc.sh [ntsc-final|pal-final|jpn-final]
 #
-# Dependencies (see docs/internals.md):
+# Dependencies (see docs/building.md):
 #   - CMake >= 3.16
 #   - SDL2 dev
 #   - zlib dev
 #   - OpenGL dev (opengl32 on Windows, GL on Linux, OpenGL.framework on macOS)
 #
 # Example (Linux):
-#   sudo apt install cmake libsdl2-dev zlib1g-dev libgl4-mesa-dev
+#   sudo apt install cmake libsdl2-dev zlib1g-dev libgl1-mesa-dev
 # Example (macOS):
 #   brew install cmake sdl2 zlib
 # Example (Windows/MSYS2):
diff --git a/docs/SetupGuide.md b/docs/SetupGuide.md
index 1688aa53..7b5177b0 100644
--- a/docs/SetupGuide.md
+++ b/docs/SetupGuide.md
@@ -88,9 +88,23 @@ This project can compile the ROM's C code in one of two ways, controlled by the
   [qemu-irix](https://github.com/n64decomp/qemu-irix/releases) user-mode emulator instead. Slower, but useful as a
   lower-risk reference if you ever suspect the recompiled toolchain is misbehaving.
 
+> **The IRIX IDO binaries are not distributed with this repository.** They are
+> proprietary SGI software and only ever ran on IRIX/MIPS. Neither `tools/irix/`
+> nor `tools/SGIImageViewer.exe` is committed here; the **PC port does not need
+> them at all** (asset extraction uses `tools/extractor`, and the port itself
+> builds with CMake + your host compiler). You only need the IDO toolchain to
+> build the actual **N64 ROM** or to regenerate `tools/ido5.3_recomp`. To get it:
+> obtain the IDO 5.3 tree from an IRIX 5.3 installation, from
+> [`decompals/ido-static-recomp`](https://github.com/decompals/ido-static-recomp)
+> (which documents where its input binaries come from), or from the N64
+> decompilation community, and place it at `tools/irix/root/` (with `cc` etc.
+> under `tools/irix/root/usr/bin/`). `tools/SGIImageViewer.exe` is an optional
+> convenience only.
+
 The recompiled toolchain isn't checked into git — it's generated locally the first time it's needed, from the
-original IDO binaries in `tools/irix/root`. A plain `make` from the repo root builds it automatically if it's
-missing (wired up via `tools/Makefile`), so most people never need to run this step by hand. To build it explicitly
+original IDO binaries in `tools/irix/root` (see the note above on obtaining those). A plain `make` from the repo
+root builds it automatically if it's missing (wired up via `tools/Makefile`), so most people never need to run
+this step by hand. To build it explicitly
 (e.g. to watch its own build output, or force a rebuild after changing it):
 
 ```bash
diff --git a/docs/building.md b/docs/building.md
index f72da027..85a6f61e 100644
--- a/docs/building.md
+++ b/docs/building.md
@@ -35,10 +35,13 @@ versions and hashes.
 ### Asset extraction (decompilation toolchain)
 
 The extraction scripts need `binutils-mips-linux-gnu` (or an equivalent MIPS
-binutils), `make`, `git`, `python3`, and `qemu-irix` for the IRIX IDO
-compiler. On Windows this is easiest under WSL or a Linux VM. Full details and
-alternatives (Docker, the recompiled IDO toolchain) are in
-[`SetupGuide.md`](SetupGuide.md).
+binutils), `make`, `git`, and `python3`. They build a small host-compiled
+`tools/extractor` and slice blobs straight out of the ROM — **no IDO / IRIX
+toolchain is involved in extraction or in the PC build.** (The IDO toolchain is
+only needed to build the N64 ROM itself, and its proprietary SGI binaries are
+not distributed here — see [`SetupGuide.md`](SetupGuide.md) "Recompile IDO".)
+On Windows this is easiest under WSL or a Linux VM. Full details and
+alternatives (Docker) are in [`SetupGuide.md`](SetupGuide.md).
 
 ---
 
diff --git a/docs/dev-process.md b/docs/dev-process.md
index ad368fcc..1aeefec2 100644
--- a/docs/dev-process.md
+++ b/docs/dev-process.md
@@ -40,14 +40,14 @@ more room.
 
 Lewis's agents recorded generalisable IDO quirks in a learnings file; later
 agents read it and did better. Our equivalent is
-[`porting-notes.md`](porting-notes.md) — the recurring N64->PC bug classes,
-one screen, each entry an index into the full [`dev/findings.md`](dev/findings.md)
+[`porting-notes.md`](porting-notes.md) — the recurring N64→PC bug classes,
+each entry a terse index into the full [`dev/findings.md`](dev/findings.md)
 log. Every investigation brief links it, and every investigation ends by
 appending any new generalisable quirk.
 
 The classes it currently tracks:
 
-- pointer-width struct growth (32->64) in ROM-serialized or pun-allocated
+- pointer-width struct growth (32→64) in ROM-serialized or pun-allocated
   structs — the dominant class;
 - 16-byte host `Gfx`/`Vtx` vs 8-byte N64 — reservation and copy sizing;
 - big-endian rodata read on a little-endian host: `f32` word-pairs, header
diff --git a/docs/dev/AUDIO-PLAN.md b/docs/dev/AUDIO-PLAN.md
index eeb4da6a..8f12da69 100644
--- a/docs/dev/AUDIO-PLAN.md
+++ b/docs/dev/AUDIO-PLAN.md
@@ -7,6 +7,24 @@ for closing **D77** (no audible music/sound on PC). Companion docs: `docs/HANDOF
 
 ---
 
+## Contents
+
+1. [Goal and scope](#1-goal-and-scope)
+2. [Current state](#2-current-state-verified-by-reading-not-assumption)
+3. [The GE pipeline in one paragraph](#3-the-ge-pipeline-in-one-paragraph)
+4. [How the PD port does it](#4-how-the-pd-port-does-it-ground-truth-comparison)
+5. [Architecture options](#5-architecture-options-evaluate-all-recommendation-marked) — options A/B/C + sub-decisions A1–A3
+6. [DMEM layout](#6-dmem-layout-ido-from-srclibultraaudiosynthinternalsh5771)
+7. [Shims and ABI fixes required](#7-shims-and-abi-fixes-required) — incl. [7.1 `K0_TO_PHYS` bug](#71-bug-found-k0_to_phys-is-unshimmed-and-corrupts-pointers-on-pc)
+8. [Phase 0 — recon](#8-phase-0--recon-do-first-informs-everything)
+9. [Open questions](#9-open-questions-decide-before-phase-1)
+10. [Phases and milestones](#10-phases-and-milestones)
+11. [Verification ritual (audio-specific)](#11-verification-ritual-audio-specific-on-top-of-the-standard-one-in-agentsmd)
+12. [Risks / unknowns](#12-risks--unknowns)
+13. [Key references](#13-key-references)
+
+---
+
 ## 1. Goal and scope
 
 Make GoldenEye's audio audible on the PC port with 1:1 fidelity to the N64 pipeline:
diff --git a/docs/dev/GRAPHICS-BACKLOG.md b/docs/dev/GRAPHICS-BACKLOG.md
index 18e6c819..12a6a22f 100644
--- a/docs/dev/GRAPHICS-BACKLOG.md
+++ b/docs/dev/GRAPHICS-BACKLOG.md
@@ -16,6 +16,9 @@ once the game is playable start-to-finish without crashes.
 | ~~**D116 (HUD)**~~ | ~~proportional-font strings + ammo counter mirrored~~ | **CLOSED — NOT A BUG (M-33, §F D168).** Same capture-orientation artifact as D114. |
 | **D148** | Level-end cinematics don't play. Dam: hitting the exit trigger should show Bond rappelling down the dam (the bungee jump); instead it cuts straight to the mission-complete report. | Scripted cutscene / cinematic-camera + Bond rappel anim. Found M-27 during the first full Dam→Facility playthrough. Progression is unaffected — the report and next briefing load fine. **M-31 (§F D160):** static-traced the Dam end ailists (`ai_24`→`ai_17`, `camera_switch`→`CutsceneRecord` type-46 lookup). Diagnostic `GE_D160=1` shipped (`c95713f5`) — user runs Dam to the exit with it set; the trace pins whether the cutscene AI runs, and whether the propDef-index walk (D122/D132 family) hands `camera_switch` a bogus cutscene record. Leading hypothesis: propDef `sizepropdef` stride desync (medium confidence), not D75 anim-model. |
 | **D149** | Front-end MISSION COMPLETE / mode-select 3D models (dossier, wallets) render as garbled full-screen geometry or not at all — one of the compiled sub-DLs (seg5+0x9ee4) is malformed, D146 now skips it instead of aborting. | D75 family. §F D146. |
+| **D170** | Named-NPC "flee" scripted behaviour looks wrong on PC: on Silo, Ourumov breaks into a full run the instant he alerts ("Ourumov: Kill him!") rather than the N64 walk/retreat; developer reports Trevelyan behaves the same way on Cradle. | **Observed, not investigated** (M-34, from a Silo playtest capture; user bug report — no fix wanted yet). Could be a genuine ailist / `chrTick` divergence, OR the D155/D156 wall-clock `deltaFrames` locomotion-anim speed-up (a catch-up spike playing the walk cycle at 2–3×). **Verify first:** capture the same beat against N64 reference footage before assuming it diverges — Ourumov may be scripted to flee. If it's real, suspect the frametiming clamp (§F D155/D156) or the Cradle-family AI path (§F D123). |
+| **D171** | Silo missile-shaft: a large translucent grey triangle sweeps across the lower frame while panning down the shaft (~3 s of the capture). Looks like a stray poly / near-plane clip fan, same family as D106 (portal near-plane min>max). | **Observed, not investigated** (M-34, Silo capture). Cosmetic, intermittent, tied to camera angle in the open vertical shaft. |
+| **D172** | Bullet-impact / blood particle spray renders magenta or cyan instead of dark red (visible on guards hit at close range on Silo). | **Observed, not investigated** (M-34, Silo capture). A channel-order / CC-mode error in the particle-sprite path (magenta = R+B, cyan = G+B) — not the D64/D120 blood-decal family (those are crash/hang + missing decals, not colour). Likely a texture-format or env-colour decode in `blood_animation.c` / the fast3d sprite path. |
 
 ## Resolved (for reference)
 
diff --git a/docs/dev/HANDOFF-ARCHIVE.md b/docs/dev/HANDOFF-ARCHIVE.md
index cb51a5bb..109fb7e0 100644
--- a/docs/dev/HANDOFF-ARCHIVE.md
+++ b/docs/dev/HANDOFF-ARCHIVE.md
@@ -1,8 +1,9 @@
 # HANDOFF ARCHIVE — session-by-session narrative
 
-> **Reference only.** Current state, next task, and environment live in
-> `docs/HANDOFF.md`; per-finding detail in `docs/internals.md` §F/§H
-> (indexed at the top of §F). This file is the frozen accumulation of
+> **Reference only.** Current state lives in the [README](../../README.md)
+> "Status" section and `docs/dev/LEVEL-STATUS.md`; per-finding detail is in
+> [`findings.md`](findings.md) §F/§H (indexed at the top of §F). This file
+> is the frozen accumulation of
 > prior sessions' handoff briefs (through session M-11), kept so the
 > reasoning behind a fix is recoverable. Everything below is as-written
 > at the time and may be superseded.
diff --git a/docs/dev/agentic-development.md b/docs/dev/agentic-development.md
index 76c04bf2..5c61e2a4 100644
--- a/docs/dev/agentic-development.md
+++ b/docs/dev/agentic-development.md
@@ -1,4 +1,19 @@
-# Agentic development: goal, method, timeline, assessment
+# Agentic development: a two-agent experiment
+
+*A GoldenEye 007 N64→desktop port, taken from "builds a ROM" to "playable front
+end" in about two weeks — driven mostly by two coding agents, a local
+open-weight model and a hosted frontier model, handing work back and forth
+through shared written notes under one person's direction. This is the goal,
+the method, the timeline, and an honest read on what did and didn't work.*
+
+## Contents
+
+- [Why this project exists](#why-this-project-exists)
+- [The setup](#the-setup)
+- [Timeline](#timeline)
+- [By the numbers](#by-the-numbers) — [Commit velocity](#commit-velocity) · [Who did what](#who-did-what)
+- [The handoff workflow](#the-handoff-workflow)
+- [Assessment](#assessment)
 
 ## Why this project exists
 
@@ -26,10 +41,10 @@ and forth** through shared written artifacts, directed by one human.
 
 | | |
 |---|---|
-| Local agent | `unsloth/Qwen3.8-27B-GGUF:UD-Q4_K_XL` on a single **NVIDIA RTX 5090**, driven mainly through the **[pi](https://pi.dev/)** coding agent. Unsloth Desktop (a newer local agentic-coding harness) was also trialed but not used significantly. |
+| Local agent | `unsloth/Qwen3.8-27B-GGUF:UD-Q4_K_XL` on a single **NVIDIA RTX 5090**, driven mainly through the **[pi](https://pi.dev/)** coding agent. Unsloth Desktop (Unsloth's local model runtime, which can drive agents such as Claude Code) was also trialed but not used significantly. |
 | Hosted agent | **Claude**, via **Claude Code**, on a Claude Pro subscription — mostly **Sonnet 5**, with **Opus 5** used as an escalation tier for the hardest problems and whenever there was subscription budget to spend on it |
 | Human | one person: direction, work partitioning, integration, and every build / playtest / frame-capture the agents could not run |
-| Base | fork of the [GoldenEye 007 decompilation](https://github.com/n64decomp/007) (years of prior work by Larry Ficken / "KholdFuzion" and contributors) |
+| Base | fork of the [GoldenEye 007 decompilation](https://github.com/n64decomp/007) (years of prior work by Larry Ficken ("kholdfuzion") and contributors) |
 | Reused | the [Perfect Dark PC port](https://github.com/fgsfdsfgs/perfect_dark)'s `fast3d` software RSP — same Rare engine family |
 
 The port is the GoldenEye-specific porting work **on top of** those two
@@ -56,13 +71,13 @@ gantt
     OS shims, threads, fast3d integration :2026-08-21, 2d
     First rendered frames                 :m4, 2026-08-22, 1d
     Offline asset-conversion pipeline     :2026-08-22, 2d
-    Entire intro renders (logos->cast)    :m5, 2026-08-24, 1d
+    Entire intro renders (logos to cast)  :m5, 2026-08-24, 1d
     section Both agents - handoff workflow
     Claude joins                          :milestone, 2026-08-27, 0d
     Stage load; Bunker 1 renders + firefight :2026-08-27, 2d
-    21 solo levels: load + render + no-crash :m6, 2026-08-29, 1d
+    21 solo levels load + render + no-crash  :m6, 2026-08-29, 1d
     SDL input layer (kbd/mouse/gamepad)   :2026-08-29, 1d
-    Front-end flow menu->briefing->start  :m7, 2026-08-30, 1d
+    Front-end flow (menu to briefing to start) :m7, 2026-08-30, 1d
     File-backed EEPROM saves              :2026-08-31, 1d
 ```
 
@@ -75,26 +90,22 @@ gantt
 - **Day 11** (27 Aug): second agent joins.
 - **Day 13** (29 Aug): **all 21 solo missions load, render, and survive an
   unattended play window without crashing.**
-- **Day 14** (30–31 Aug): front end playable end to end (menu -> mission
-  select -> difficulty -> briefing -> start); file-backed saves.
+- **Day 14** (30–31 Aug): front end playable end to end (menu → mission
+  select → difficulty → briefing → start); file-backed saves.
 
 So roughly **two weeks**, one person part-time, to take a decompilation from
 "builds an N64 ROM" to "boots on desktop, renders every solo level, playable
 through the front end into the early game" — with audio and a set of cosmetic
 issues still outstanding.
 
-We are not aware of a prior Nintendo 64 native PC port developed primarily
-through coding agents, or of one reaching this state on this timeline. That is
-a statement about how little precedent there is, not a claim of a record.
-
 ## By the numbers
 
 | | |
 |---|---|
 | Calendar time | 16 days (16 Aug – 1 Sep 2026), one person part-time |
-| Commits on the port | 217 |
-| Root-caused bugs logged | 160 (`D1`–`D167` in [`findings.md`](findings.md)) |
-| Handoff sessions | ~30 (`M-2` … `M-32`) |
+| Commits on the port | ~223 |
+| Root-caused bugs logged | 162 (`D1`–`D169` in [`findings.md`](findings.md); some later merged or withdrawn) |
+| Handoff sessions | ~33 (`M-2` … `M-33`) |
 | Game-source files given `#ifdef PORT` ABI edits | 63 files, 241 blocks |
 | New port-layer / tooling files | 128 |
 | Port layer | ~17,000 lines C/C++ (`port/`) |
@@ -108,20 +119,24 @@ xychart-beta
     title "Commits per day"
     x-axis ["8/16", "8/17", "8/20", "8/21", "8/22", "8/23", "8/24", "8/27", "8/28", "8/29", "8/30", "8/31", "9/1"]
     y-axis "commits" 0 --> 60
-    bar [1, 2, 2, 6, 9, 2, 4, 7, 57, 39, 39, 46, 3]
+    bar [1, 2, 2, 6, 9, 2, 4, 7, 57, 39, 39, 46, 9]
 ```
 
-The 8/25–8/26 gap is between phases; the jump on 8/28 is the second agent
-joining, plus the parallel multi-agent "bursts" used near the end.
+Days with no commits are omitted. The 8/25–8/26 gap is between phases; the
+step up from 8/28 onward is the collaborative phase in full swing, including
+the parallel multi-agent "bursts" used near the end.
 
 ### Who did what
 
 ```mermaid
-pie showData title "Commits by phase"
-    "Phase A - local model, solo (16-24 Aug)" : 26
-    "Phase B - collaborative (27 Aug-1 Sep)" : 191
+pie showData title "Commits by agent (raw count)"
+    "Claude" : 161
+    "Local model (Qwen 3.8 / pi)" : 62
 ```
 
+(Phase A — the first 26 commits, to 24 Aug — was entirely the local model,
+solo; from 27 Aug on the two agents worked in parallel.)
+
 | Milestone / workstream | Primary agent | Weight |
 |---|---|---|
 | PC build system, region macros, full compile + link of ~230 units | local model | large |
@@ -130,7 +145,7 @@ pie showData title "Commits by phase"
 | Offline asset-conversion architecture + first converters (`tools_pc/`) | local model | large |
 | Intro rendering (logos → gun-barrel → cast) | local model | medium |
 | Stage load unblocked (`D69`–`D87`) | Claude | medium |
-| 21-level crash sweep — ~12 crash classes root-caused (`D88`–`D167`) | Claude | large |
+| 21-level crash sweep — ~12 crash classes root-caused (`D88`–`D169`) | Claude | large |
 | SDL input layer (keyboard / mouse / gamepad, mouse-look) | Claude | medium |
 | Front-end flow (menu → briefing → start), EEPROM saves | Claude | medium |
 | Parallel struct-layout / converter static audits | local model | medium |
@@ -138,14 +153,15 @@ pie showData title "Commits by phase"
 | The hardest structural bugs (matrix handedness, format specs) | Claude | — |
 | Every build, playtest, frame capture, integration, and direction | human | — |
 
-**Estimated effort split.** By raw commit count it is ~28% local model /
-~72% Claude (Phase A entirely local, ~15–20% of Phase B local). Weighting by
-milestone difficulty rather than commit count — the foundation the local model
-built is a heavier third of the project than its commit count suggests — the
-developer's estimate is roughly:
+**Estimated effort split.** The raw commit count above (~28% local / ~72%
+Claude) undercounts the local model: Phase A landed the build system and boot
+chain in relatively few, large commits, and the local model kept contributing
+~15–20% of Phase B in parallel. Weighting by milestone difficulty rather than
+raw commits — the foundation is a heavier third of the project than its commit
+share suggests — the developer's estimate is roughly:
 
 ```mermaid
-pie showData title "Estimated agent effort (milestone-weighted)"
+pie showData title "Agent effort, milestone-weighted"
     "Claude" : 60
     "Local model (Qwen 3.8 / pi)" : 40
 ```
@@ -163,8 +179,8 @@ This is the part worth paying attention to.
 
 ```mermaid
 flowchart LR
-    H[Human: direction, integration,<br/>build + playtest verification]
-    C[Claude / Claude Code<br/>frontier, hosted]
+    H["Human: direction, integration,<br/>build + playtest verification"]
+    C["Claude / Claude Code<br/>frontier, hosted"]
     Q["Qwen 3.8 via pi<br/>open-weight, local RTX 5090"]
     D[(Shared artifacts:<br/>HANDOFF.md · findings.md · porting-notes.md)]
 
@@ -172,10 +188,10 @@ flowchart LR
     H -->|scopes task, budget, files| Q
     C <-->|reads / appends| D
     Q <-->|reads / appends| D
-    C -.->|usage limit reached ->| Q
-    Q -.->|hard structural bug ->| C
-    C -->|patch + writeup| H
-    Q -->|patch + writeup| H
+    C -.->|usage limit reached| Q
+    Q -.->|hard structural bug| C
+    C -->|patch + write-up| H
+    Q -->|patch + write-up| H
 ```
 
 Both agents worked against the **same three written artifacts**, which is what
@@ -187,12 +203,12 @@ let them substitute for each other:
    usage limit mid-problem, the local model picked the task up from the
    HANDOFF state and continued; when the local model hit a bug that needed
    deeper structural reasoning, it wrote up where it was and Claude took over.
-2. **`findings.md`** — the chronological finding log. 160 numbered entries,
+2. **`findings.md`** — the chronological finding log. 162 numbered entries,
    each a root cause with `file:line` evidence and the fix. New agents (either
    model) are pointed at the relevant entries before they start.
 3. **`porting-notes.md`** — the append-only "recurring bug classes" file. The
    single highest-leverage artifact: it stopped both models from
-   re-deriving the same class of N64->PC bug over and over.
+   re-deriving the same class of N64→PC bug over and over.
 
 Around these, the working rules (full detail in
 [`../dev-process.md`](../dev-process.md)):
@@ -200,7 +216,7 @@ Around these, the working rules (full detail in
 - **file-partitioned tasks** — each agent's task scoped to a disjoint set of
   files so patches never collided;
 - **visible budgets** — every investigation task carried an explicit
-  "N build->run cycles" limit and a defined fallback (revert probes, write up
+  "N build→run cycles" limit and a defined fallback (revert probes, write up
   with a confidence rating);
 - **the human owns verification** — building, running the game, capturing a
   frame, and judging it against N64 reference footage was never delegated.
diff --git a/docs/img/archives-1.png b/docs/img/archives-1.png
deleted file mode 100644
index 96052755..00000000
Binary files a/docs/img/archives-1.png and /dev/null differ
diff --git a/docs/img/archives-2.png b/docs/img/archives-2.png
deleted file mode 100644
index 806909d6..00000000
Binary files a/docs/img/archives-2.png and /dev/null differ
diff --git a/docs/img/attract-silo.png b/docs/img/attract-silo.png
deleted file mode 100644
index 21666ff2..00000000
Binary files a/docs/img/attract-silo.png and /dev/null differ
diff --git a/docs/img/bunker1-1.png b/docs/img/bunker1-1.png
deleted file mode 100644
index 682726a3..00000000
Binary files a/docs/img/bunker1-1.png and /dev/null differ
diff --git a/docs/img/bunker1-2.png b/docs/img/bunker1-2.png
deleted file mode 100644
index dc7c7a0c..00000000
Binary files a/docs/img/bunker1-2.png and /dev/null differ
diff --git a/docs/img/bunker1-3.png b/docs/img/bunker1-3.png
deleted file mode 100644
index 4a6c99ae..00000000
Binary files a/docs/img/bunker1-3.png and /dev/null differ
diff --git a/docs/internals.md b/docs/internals.md
index 767c7c1b..3e07f7db 100644
--- a/docs/internals.md
+++ b/docs/internals.md
@@ -17,6 +17,27 @@ platforms.
 
 ---
 
+## Contents
+
+1. [Executive summary](#1-executive-summary)
+2. [What the PD port actually does (the reference architecture)](#2-what-the-pd-port-actually-does-the-reference-architecture)
+   — [2.1 Emulate the RSP, bypass the RDP](#21-the-key-architectural-insight-emulate-the-rsp-bypass-the-rdp)
+   · [2.2 How the RSP is invoked](#22-how-the-rsp-is-invoked)
+   · [2.3 Build system](#23-build-system)
+   · [2.4 PD-port copy-candidate audit](#24-pd-port-copy-candidate-audit-session-2026-08-22)
+3. [GoldenEye decomp: what we're porting](#3-goldeneye-decomp-what-were-porting)
+   — [3.1 Entry / main loop](#31-entry--main-loop)
+   · [3.2 The two libultra trees](#32-the-two-libultra-trees-srclibultra-and-srclibultrare)
+4. [N64 hardware surfaces that must be shimmed](#4-n64-hardware-surfaces-that-must-be-shimmed)
+5. [GE-specific graphics differences vs PD (the RSP work)](#5-ge-specific-graphics-differences-vs-pd-the-rsp-work)
+6. [Audio](#6-audio)
+7. [Input & saves](#7-input--saves)
+8. [Recommended plan](#8-recommended-plan) — phases 0–4
+9. [Open questions / risks](#9-open-questions--risks)
+10. [References](#10-references)
+
+---
+
 ## 1. Executive summary
 
 * **Feasibility: HIGH.** GoldenEye and Perfect Dark share the same Rare
diff --git a/docs/media/goldeneye-demo.gif b/docs/media/goldeneye-demo.gif
new file mode 100644
index 00000000..b5a6237a
Binary files /dev/null and b/docs/media/goldeneye-demo.gif differ
diff --git a/docs/porting-notes.md b/docs/porting-notes.md
index 9617910c..0e4aeaa1 100644
--- a/docs/porting-notes.md
+++ b/docs/porting-notes.md
@@ -1,13 +1,29 @@
 # Porting notes — recurring N64→PC bug classes
 
-A one-screen field guide to the bug classes that keep recurring when running
-big-endian 32-bit N64 game code, unmodified, on a little-endian 64-bit host.
-Each entry cites a `Dxx` label; the full evidence and fix for that instance
-live in [`dev/findings.md`](dev/findings.md) under the same label.
+A field guide to the bug classes that keep recurring when running big-endian
+32-bit N64 game code, unmodified, on a little-endian 64-bit host. Terse by
+design — each entry compresses a full investigation to a symptom, a fix, and a
+grep heuristic for finding siblings. Each entry cites a `Dxx` label; the full
+evidence and fix for that instance live in [`dev/findings.md`](dev/findings.md)
+under the same label. Skim the section headers; read the classes relevant to
+the task at hand.
 
 If you are debugging a crash in this port, read this first — the odds are
 good that you are looking at one of these.
 
+## Contents
+
+- [A. Pointer-width struct growth (32→64) — the dominant class](#a-pointer-width-struct-growth-3264--the-dominant-class)
+- [B. 16-byte PC `Gfx` / `Vtx` vs 8-byte N64](#b-16-byte-pc-gfx--vtx-vs-8-byte-n64)
+- [C. Big-endian rodata / ROM data read on little-endian PC](#c-big-endian-rodata--rom-data-read-on-little-endian-pc)
+- [C2. Port-layer / SDL shims](#c2-port-layer--sdl-shims)
+- [D. N64 hardware idioms fast3d does not emulate](#d-n64-hardware-idioms-fast3d-does-not-emulate)
+- [D2. The HUD/model "X-mirror" — RESOLVED](#d2-the-hudmodel-x-mirror-d114d116--resolved-it-was-an-upside-down-capture)
+- [D3. GCC/mingw makes an all-non-negative `enum` UNSIGNED](#d3-gccmingw-makes-an-all-non-negative-enum-unsigned)
+- [D4. N64 "interrupts off" must be a real lock on PC](#d4-n64-interrupts-off-is-not-free-on-pc--it-must-be-a-real-lock)
+- [D5. Loop bounds that assume linker adjacency of two globals](#d5-loop-bounds-that-assume-linker-adjacency-of-two-file-scope-globals)
+- [E. Process / method notes](#e-process--method-notes)
+
 ## A. Pointer-width struct growth (32→64) — the dominant class
 
 A decomp struct with pointer fields, or one pun-allocated into a fixed
diff --git a/scripts/__pycache__/generate_prop_model_c.cpython-312.pyc b/scripts/__pycache__/generate_prop_model_c.cpython-312.pyc
deleted file mode 100644
index 55ec2665..00000000
Binary files a/scripts/__pycache__/generate_prop_model_c.cpython-312.pyc and /dev/null differ
diff --git a/scripts/__pycache__/generate_prop_model_c.cpython-38.pyc b/scripts/__pycache__/generate_prop_model_c.cpython-38.pyc
deleted file mode 100644
index c7d78378..00000000
Binary files a/scripts/__pycache__/generate_prop_model_c.cpython-38.pyc and /dev/null differ