feat: alerts (M8), notifications config page, and app icon (0.4.0)

- feat(alerts): desktop notifications (notify-send) for overheat (GPU/CPU past a
  configurable threshold), GPU-lost, and a new-version-available alert (once per
  version). Edge-triggered with cooldown so it doesn't spam (core/alerts.py)
- feat(gui): Notifications page to configure alerts (enable, GPU/CPU thresholds,
  Send test); changes apply live and persist via config.save_config/update_config
- feat(gui): ship a RigDoctor icon; the GUI self-registers the icon + .desktop on
  launch and sets the Wayland app-id, so the dock shows it after an update + relaunch
  (no installer re-run); installer/uninstaller updated to manage the icon
- config: alerts_enabled, gpu_temp_alert, cpu_temp_alert; flat-TOML writer
- tests for the alert monitor and config round-trip

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

.gitea/workflows/release.yml

@@ -27,12 +27,39 @@ jobs:
           python -m pip install --upgrade build
           python -m build
 
+      - name: Build self-extracting installer (.run)
+        run: python packaging/make_run.py
+
       - name: Read version
         id: ver
         run: |
           V=$(python -c "import tomllib; print(tomllib.load(open('pyproject.toml','rb'))['project']['version'])")
           echo "version=$V" >> "$GITHUB_OUTPUT"
 
+      - name: Build release notes
+        run: |
+          python - <<'PY'
+          import json
+          version = "${{ steps.ver.outputs.version }}"
+          tag = f"v{version}"
+          out, capturing = [], False
+          try:
+              for line in open("CHANGELOG.md", encoding="utf-8").read().splitlines():
+                  if line.startswith("## "):
+                      if capturing:
+                          break
+                      capturing = line.startswith(f"## [{version}]")
+                      continue
+                  if capturing:
+                      out.append(line)
+          except OSError:
+              pass
+          body = "\n".join(out).strip() or f"Release {tag}."
+          payload = {"tag_name": tag, "target_commitish": "${{ github.sha }}", "name": tag, "body": body}
+          open("/tmp/release.json", "w", encoding="utf-8").write(json.dumps(payload))
+          print(f"release notes: {len(body)} chars")
+          PY
+
       - name: Publish Gitea release
         env:
           TOKEN: ${{ secrets.GITHUB_TOKEN }}
@@ -52,7 +79,7 @@ jobs:
           rid=$(curl -sS -X POST \
             -H "Authorization: token ${TOKEN}" \
             -H "Content-Type: application/json" \
-            -d "{\"tag_name\":\"${TAG}\",\"target_commitish\":\"${{ github.sha }}\",\"name\":\"${TAG}\",\"body\":\"Automated release for ${TAG}. See CHANGELOG.md.\"}" \
+            -d @/tmp/release.json \
             "${API}/releases" | python -c "import sys, json; print(json.load(sys.stdin)['id'])")
 
           for f in dist/*; do

CHANGELOG.md

@@ -5,6 +5,133 @@ All notable changes to RigDoctor are recorded here. Format follows
 (`MAJOR.MINOR.PATCH`, pre-1.0). `__version__` and `pyproject.toml` must match the git
 release tag (so the auto-updater, D18, can compare versions).
 
+## [0.4.0] - 2026-05-21
+### Added
+- **Alerts (M8)**: desktop notifications (via `notify-send`) for **overheat** (GPU/CPU past a
+  threshold), **GPU-lost** (nvidia-smi timeout), and a **new version available** (fired once
+  per version). Edge-triggered with a cooldown so it doesn't spam. Degrades gracefully if
+  `notify-send` isn't installed.
+- **Notifications page**: configure alerts (enable/disable, GPU/CPU temperature thresholds)
+  with a "Send test" button; changes apply live and persist to `config.toml`.
+- **App icon**: ships a RigDoctor icon and shows it in the dock/launcher. The GUI
+  **self-registers** the icon + `.desktop` on launch (and sets the Wayland app-id), so a
+  self-update + relaunch picks it up — no need to re-run the installer.
+
+## [0.3.2] - 2026-05-21
+### Changed
+- Replaced the per-page "Run with admin" buttons with a **single password prompt at launch**
+  (`pkexec`): the GUI collects root-only data (SMART + dmidecode board/BIOS/RAM) once and
+  caches it for the session, so Health and Inventory always show the full picture. Falls back
+  to non-root if cancelled/unavailable; disable via `elevate_on_launch = false`.
+
+## [0.3.1] - 2026-05-21
+### Fixed
+- Changelog/release notes now **render Markdown** instead of showing raw `#`/`**` markup —
+  the in-app changelog uses `QTextEdit.setMarkdown()` and the update prompt renders notes as
+  rich text (closes #1).
+
+## [0.3.0] - 2026-05-21
+### Added
+- **System inventory (M5)**: CPU, GPU (model/driver/VBIOS/VRAM/PCIe), motherboard/BIOS, RAM
+  (total + modules), storage, kernel, and display server. CLI `rigdoctor inventory`
+  (`--json` / `--markdown` / `--output`) and a GUI **Inventory** tab with Copy-as-Markdown,
+  Save, and "Run with admin" (for `dmidecode` board/BIOS/RAM details). Fills the last GUI tab.
+
+## [0.2.0] - 2026-05-21
+### Added
+- **"Check for updates" button** in the sidebar — force an immediate version check instead of
+  waiting for the 30-minute poll.
+
+## [0.1.1] - 2026-05-21
+### Fixed
+- Dialogs (the update prompt and changelog) were light-on-light and unreadable — they now use
+  the dark theme with readable text.
+
+## [0.1.0] - 2026-05-21
+_First milestone release — a complete, installable, self-updating RigDoctor: live monitoring,
+crash capture + health report, desktop GUI, user-local install/uninstall, and updates._
+### Added
+- **In-app uninstaller**: "Uninstall RigDoctor" button on the Setup page (and
+  `rigdoctor uninstall [--purge]`) — removes the venv, launchers, and desktop entry, with an
+  option to also wipe settings/token/logs. Runs detached so it can delete its own venv.
+- **In-app changelog**: a "Changelog" link in the sidebar opens the release history (tags +
+  notes) fetched from the update server.
+
+## [0.0.10] - 2026-05-21
+### Added
+- **"Restart now" button** after a successful in-app update — relaunches RigDoctor for you
+  instead of asking you to restart manually.
+- **Real release notes**: CI now sets each Gitea release's body from the matching CHANGELOG
+  section (instead of "Automated release for…"), and the updater shows **"What's new"** — a
+  notes dialog before applying (GUI) and in `rigdoctor update` (CLI).
+### Changed
+- Setup page / `rigdoctor install`: dropped internal module references (M4, M5, …) from the
+  component descriptions — end users don't need them.
+- Adopting **Conventional Commits** + **git-cliff** (`cliff.toml`, `packaging/changelog.sh`)
+  to generate CHANGELOG entries from commit history going forward (D20).
+### Fixed
+- The self-extracting **`.run` installer** is now built **without makeself** (a pure-Python
+  self-extractor, `packaging/make_run.py`), so it reliably attaches to every release — it was
+  silently skipped before because the CI runner couldn't install makeself.
+
+## [0.0.8] - 2026-05-21
+### Added
+- **Periodic update checks**: the GUI now re-checks for new releases while running (every
+  `update_check_minutes`, default 30; 0 disables), so a newly published version is detected
+  without restarting. After applying an update, re-checks stop until restart.
+- **"Run with admin" on the Health page**: runs all checks (including root-only SMART) via
+  `pkexec rigdoctor report --json`, so the full report — not just "SMART needs root" — is
+  available from the UI.
+
+## [0.0.7] - 2026-05-21
+### Added
+- **User-local installer** `install.sh` (no root): creates a private venv, links
+  `rigdoctor`/`rigdoctor-gui` into `~/.local/bin`, and adds a desktop entry. Re-run to
+  upgrade; `--uninstall` to remove.
+- **Self-extracting `.run` installer** via `packaging/make-run.sh` (makeself) — one
+  download-and-run executable bundling the wheel + `install.sh`; built and attached to each
+  release by CI.
+- **Self-update apply (M13)**: `rigdoctor update` now installs the newer version via
+  authenticated pip (`rigdoctor[gui] @ git+https://oauth2:<token>@…@<tag>`); the GUI sidebar
+  "Update to v…" button applies it and prompts to restart. Token is scrubbed from output.
+
+## [0.0.6] - 2026-05-21
+### Added
+- **Token-gated updates (M13)**: store a Gitea Personal Access Token, **encrypted in the OS
+  keyring** (Secret Service / GNOME Keyring via `secret-tool`) with a 0600-file fallback.
+  `rigdoctor login` / `logout` / `update [--check]`; GUI **Setup → Update access** panel
+  (token field, "Get a token", backend status) and sidebar states (connect / up-to-date /
+  "Update to v…" / access denied). Updates are gated to accounts on the Gitea server (D18).
+- `libsecret-tools` added to the installer catalog (enables encrypted token storage).
+### Changed
+- D18 update mechanism revised from anonymous public HTTP to **authenticated HTTP (token)** —
+  the Gitea instance requires sign-in for all anonymous access.
+
+## [0.0.5] - 2026-05-21
+### Added
+- **M9 installer (first cut)**: detects distro / package manager / GPU; a catalog of optional
+  components (smartmontools, lm-sensors, dmidecode, pciutils, libnotify) with what each
+  enables; `rigdoctor install [--check] [-y]` installs missing apt packages via pkexec/sudo
+  with consent; GUI **Setup** tab with one-click install. Fixes the "smartmontools missing"
+  gap in the health report.
+- **Update check (M13, check half)**: on GUI launch the sidebar checks the Gitea releases API
+  and shows "up-to-date", an "Update to v…" button if a newer release exists, or "update check
+  unavailable" if the API can't be reached anonymously.
+
+## [0.0.4] - 2026-05-21
+### Added
+- **M4 health report**: scans kernel logs (NVIDIA Xid incl. 79 "fell off the bus", kernel
+  panic, OOM, MCE, PCIe AER, thermal, amdgpu reset), SMART health, NVIDIA driver/library
+  mismatch, journald persistence, and live temps → prioritized plain-language findings with
+  suggested fixes (read-only, D9).
+- CLI `rigdoctor report` (text + `--json`).
+- GUI **Health** tab: runs checks in the background; findings shown as severity-colored cards.
+- Tests for the journal scanner.
+
+## [0.0.3] - 2026-05-21
+### Added
+- Show the app version (`v<version>`) in the GUI sidebar.
+
 ## [0.0.2] - 2026-05-21
 ### Added
 - **M3 crash-capture logger**: crash-safe JSONL (`fsync` per sample), size-based rotation,

README.md

@@ -2,10 +2,10 @@
 
 A **modular diagnostics, monitoring, and health-check toolkit for Linux gamers.**
 
-> **Status:** 🟢 Phase 1 (MVP) in progress. The **sensor core (M1)** and **crash-capture
-> logger (M3)** work — `snapshot`/`monitor` read NVIDIA GPU, CPU, memory, and NVMe live, and
-> `record` captures a crash-safe log with a post-crash report. A desktop GUI (M10) is also
-> up. Health report (M4) is next. See `docs/ROADMAP.md`.
+> **Status:** 🟢 Phase 1 (MVP) complete. The **sensor core (M1)**, **crash-capture logger
+> (M3)**, and **health report (M4)** all work — live `snapshot`/`monitor`, crash-safe `record`
+> with a post-crash report, and `report` to scan logs/SMART/driver for likely causes. A
+> desktop GUI (M10) ties them together (dashboard, recording, health). See `docs/ROADMAP.md`.
 
 ## Why this exists
 
@@ -63,6 +63,21 @@ Full rationale and the still-open questions are in `docs/DECISIONS.md`.
 | `installer/` | Installer / `.deb` packaging (empty until Phase 4) |
 | `tests/` | Tests (stdlib `unittest`) |
 
+## Install (user-local, no root)
+
+RigDoctor installs into a private venv under `~/.local` — no root, self-updating:
+
+```bash
+./install.sh                 # from a source checkout or the self-extracting .run
+./install.sh --ref v0.0.6    # install a specific released tag (needs a token)
+./install.sh --uninstall     # remove it
+```
+
+This adds `rigdoctor` / `rigdoctor-gui` to `~/.local/bin` and a desktop entry. Each release
+also ships a one-file **`.run`** installer (download, `chmod +x`, run). Updates are gated to
+accounts on the Git server (a Personal Access Token); save one via the GUI **Setup → Update
+access** panel or `rigdoctor login`, then `rigdoctor update` (or the sidebar button).
+
 ## Run it (dev)
 
 Stdlib-only, no install needed (target is Python ≥ 3.11; tested on 3.14):
@@ -104,8 +119,8 @@ rigdoctor gui             # or: rigdoctor-gui
 It opens a dark-themed window with sidebar navigation and a **live dashboard** over the
 same sensor core — circular gauges for the headline metrics plus collapsible per-subsystem
 cards (GPU/CPU/memory/storage) with temperature-colored values (icey-blue → green → red).
-The **Logs** section is a full recording page (start/stop, live status, and the post-crash
-report); Health / Inventory are placeholders until M4 / M5 land.
+The **Logs** and **Health** sections are full pages (recording controls + post-crash report;
+and the kernel-log / SMART / driver scan). **Inventory** is a placeholder until M5 lands.
 
 Without the GUI extra, `pip install -e .` gives just the stdlib-only CLI.
 

cliff.toml

@@ -0,0 +1,42 @@
+# git-cliff configuration — generate CHANGELOG.md from Conventional Commits (D20).
+# Run via packaging/changelog.sh.
+
+[changelog]
+header = """
+# Changelog
+
+All notable changes to RigDoctor are recorded here. Format follows
+[Keep a Changelog](https://keepachangelog.com/); versioning is SemVer-style
+(`MAJOR.MINOR.PATCH`, pre-1.0). `__version__` and `pyproject.toml` must match the git
+release tag (so the auto-updater, D18, can compare versions).
+"""
+body = """
+{% for group, commits in commits | group_by(attribute="group") %}
+## {{ group | upper_first }}
+{% for commit in commits %}\
+- {{ commit.message | upper_first }}\
+{% endfor %}
+{% endfor %}
+"""
+trim = true
+
+[git]
+conventional_commits = true
+filter_unconventional = false
+commit_parsers = [
+    { message = "^feat", group = "Added" },
+    { message = "^fix", group = "Fixed" },
+    { message = "^docs", group = "Documentation" },
+    { message = "^perf", group = "Performance" },
+    { message = "^refactor", group = "Changed" },
+    { message = "^chore\\(release\\)", skip = true },
+    { message = "^chore|^build|^ci|^style|^test", group = "Internal" },
+    { message = ".*", group = "Other" },
+]
+tag_pattern = "v[0-9]*"
+sort_commits = "oldest"
+
+[bump]
+# Pre-1.0 rules (D21): feat -> minor, fix -> patch, breaking -> minor (not major).
+features_always_bump_minor = true
+breaking_always_bump_major = false

docs/DECISIONS.md

@@ -152,9 +152,22 @@ reachable from it. This **supersedes the earlier "CLI-first / terminal-first" fr
 - *No change to layering (D2):* the core, CLI, and daemon stay **stdlib-only** and must run
   without Qt. "GUI-first" is about emphasis and front-end parity, not dropping headless support.
 
-### D18 — Auto-update (M13) — *PLANNED 2026-05-21*
+### D18 — Auto-update (M13) — *PLANNED 2026-05-21; mechanism revised 2026-05-21*
 RigDoctor should **check for a newer version on launch and self-update** (new module **M13**).
-**Mechanism (chosen): user-local, no-root self-update from the public repo.**
+**Mechanism (revised): user-local, no-root self-update over authenticated HTTP (token).**
+*Why revised:* the Gitea instance requires sign-in for **all** anonymous access (repo page,
+releases feed, raw, API all 303/403 anonymously), so the original "public HTTP" plan can't
+work. Updates are therefore **gated to people with an account on the Gitea server**, which is
+desirable — access control is delegated to Gitea.
+- *Auth:* each user creates a **Personal Access Token** (scope `read:repository`); RigDoctor
+  stores it at `~/.config/rigdoctor/token` (mode 0600) or reads `RIGDOCTOR_TOKEN`. Requests
+  send `Authorization: token <PAT>`. Finer access = repo visibility/collaborators on Gitea.
+- *Check:* `GET /api/v1/repos/jessey/rigdoctor/releases/latest` with the token; compare tags.
+- *Apply:* `pip install --upgrade "git+https://oauth2:<token>@…/rigdoctor.git@<tag>"` into the
+  user-local venv, then restart (incl. the daemon). No root.
+- *States surfaced:* no-token → "connect to update server"; auth error → "access denied";
+  newer → "Update to v…"; else "up-to-date".
+- *Original (now-superseded) plan was anonymous public HTTP:*
 - *Install model (D8 revised):* primary install is **user-local** (`~/.local`), so the running
   app can replace its own files and update with **no apt, no root, no password prompt**.
 - *Check:* on launch, query the **public Gitea releases API**
@@ -178,12 +191,41 @@ RigDoctor should **check for a newer version on launch and self-update** (new mo
 PATCH for ordinary changes, MINOR for larger milestones). `__version__`
 (`rigdoctor/__init__.py`) and `pyproject.toml` are the single source of truth and **must match
 the git release tag** so the auto-updater (D18) can compare versions. Every change updates
-`CHANGELOG.md` (Keep a Changelog style). *Note:* an early placeholder `0.1.0` was corrected to
+`CHANGELOG.md` — now generated from **Conventional Commits** via git-cliff (see D20).
+*Milestone policy (pre-1.0):* **0.0.x** = early development; **0.1.0** = first complete,
+installable, self-updating release (reached 2026-05-21); **0.x.0** = each later milestone
+(AMD/Intel, unattended logger auto-start, session sharing…); **1.0.0** = broadly stable
+(multi-vendor/distro, no major caveats). PATCH (`0.x.PATCH`) for fixes/small changes. *Note:* an early placeholder `0.1.0` was corrected to
 follow the released **0.0.x** line — first release was **V0.0.1**; current is **0.0.2**.
 
+### D20 — Automated changelog & release notes — *DECIDED 2026-05-21*
+**Release notes are generated from our changes, surfaced in the auto-updater.**
+- *Release body:* CI sets each Gitea release's `body` from the matching `CHANGELOG.md`
+  section (was a hardcoded "Automated release for…"). The updater fetches the release `body`
+  and shows **"What's new"** — a dialog before applying (GUI) and in `rigdoctor update` (CLI).
+- *Generation:* adopt **Conventional Commits** (`feat:`/`fix:`/`docs:`/`chore:` …) and
+  **git-cliff** (`cliff.toml`, `packaging/changelog.sh`) to generate `CHANGELOG.md` from
+  commit history. Refines D19's "hand-write CHANGELOG" to "generate it from conventional
+  commits"; `__version__`/`pyproject.toml`/tag still the source of truth for the version.
+- *CI does not auto-commit the changelog* (avoids push loops) — it's regenerated by the dev
+  via the script when cutting a version; CI only reads the section for the release body.
+
+### D21 — Versioning rules & automation — *DECIDED 2026-05-21*
+The next version is **determined by the Conventional Commit types** since the last release
+(D20), so it can be auto-computed instead of guessed:
+- `fix:` / `perf:` → bump **PATCH**.
+- `feat:` → bump **MINOR** (pre-1.0: `0.MINOR.0`).
+- breaking (`feat!:` / `BREAKING CHANGE:`) → pre-1.0: bump **MINOR** (not major); post-1.0: MAJOR.
+- `docs:` / `chore:` / `refactor:` / `ci:` / `test:` / `style:` alone → **PATCH** (no feature release).
+- Milestone overrides by hand are allowed (e.g., jumping to `1.0.0`); see the milestone policy in D19.
+
+*Automation:* `git-cliff --bumped-version` computes the next version from history;
+`packaging/bump.sh` writes it into `__init__.py` + `pyproject.toml`. Rules live in
+`cliff.toml [bump]` (pre-1.0: `breaking_always_bump_major = false`).
+
 ## Open
 
-None currently — all tracked decisions (D1–D19) are resolved. New questions will be added
+None currently — all tracked decisions (D1–D21) are resolved. New questions will be added
 here as they arise. Remaining detail to flesh out during build: the tray's supporting-action
 set (D13), per-module apt package names, M12's tunnel/token specifics, and M13's
 update mechanism (APT repo vs. self-installed `.deb`).

docs/MODULES.md

@@ -10,16 +10,16 @@ Status: ⬜ not started · 🟦 designing · 🟨 in progress · ✅ done
 |----|--------|--------|----------|-----------|----------|--------|
 | M1 | Sensor core | Essential | none (nvidia-smi, sysfs) | all (NVIDIA first) | P0 | ⬜ |
 | M3 | Crash-capture logger | Essential | none (opt: smartmontools) | all (NVIDIA first) | P0 | 🟨 |
-| M4 | Health report (log scan) | Essential | none (opt: smartmontools) | all (NVIDIA first) | P0 | ⬜ |
+| M4 | Health report (log scan) | Essential | none (opt: smartmontools) | all (NVIDIA first) | P0 | 🟨 |
 | M2 | Live monitor (TUI) | Monitoring | none (stdlib curses) | all | P1 | ⬜ |
-| M8 | Alerting | Monitoring | libnotify (opt) | all | P2 | ⬜ |
-| M5 | System inventory | Diagnostics | none (opt: lm-sensors, dmidecode) | all | P1 | ⬜ |
+| M8 | Alerting | Monitoring | libnotify (opt) | all | P2 | 🟨 |
+| M5 | System inventory | Diagnostics | none (opt: lm-sensors, dmidecode) | all | P1 | 🟨 |
 | M6 | Gaming env checks | Diagnostics | none | all | P2 | ⬜ |
 | M10 | Desktop GUI | Desktop UI | **python3-pyside6** | all | P2 | 🟨 |
 | M11 | Tray / menu-bar applet | Desktop UI | **python3-pyside6** (+ AppIndicator on GNOME) | all | P2 | ⬜ |
-| M9 | Installer | (meta) | none | all | P1 | ⬜ |
+| M9 | Installer | (meta) | none | all | P1 | 🟨 |
 | M12 | Session sharing / remote assist | Sharing | none (Tier 3: tmate/sshx) | all | P3 | ⬜ |
-| M13 | Auto-update | (core) | none (stdlib; user-local file swap) | all | P3 | ⬜ |
+| M13 | Auto-update | (core) | none (stdlib; user-local file swap) | all | P3 | 🟨 |
 | ~~M7~~ | ~~Stress / repro~~ | — | — | — | — | ❌ dropped (D7) |
 
 ## Notes per module
@@ -36,7 +36,9 @@ Status: ⬜ not started · 🟦 designing · 🟨 in progress · ✅ done
 - **M4 Health report** — turns scattered logs into a prioritized, plain-language findings
   list with **suggested** fixes (read-only, D9). Reuses M1 for a live snapshot. Also powers
   the **guided diagnostic session** (with M3): pick a game → focused capture → scan →
-  findings (see SPEC §4).
+  findings (see SPEC §4). *Implemented:* journalctl scan (Xid/panic/OOM/MCE/AER/thermal/amdgpu),
+  SMART, NVIDIA driver-mismatch, journald-persistence + live-temp checks; `rigdoctor report`
+  (text/JSON) + GUI Health tab. GPU-firmware verification deferred.
 - **M2 Live monitor** — depends on M1; the terminal "HWMonitor for Linux" face. Stdlib-only.
 - **M5 / M6 Diagnostics** — inventory export + gaming-env checks; M6 flags risky settings and
   suggests the fix command but does not apply it (D9).
@@ -52,14 +54,30 @@ Status: ⬜ not started · 🟦 designing · 🟨 in progress · ✅ done
   action (the guided diagnostic session), plus Open dashboard / Start-Stop recording /
   Snapshot / Quit (D13). Optional; shares the Qt dependency with M10.
 - **M9 Installer** — interactive wizard layered on the `.deb` (D8); apt-first dependency
-  resolution; enables the logger service and trigger mode.
+  resolution; enables the logger service and trigger mode. *Implemented (first cut):* distro/
+  package-manager/GPU detection (`core/sysenv`), an optional-component catalog (`core/catalog`),
+  and dependency install via pkexec/sudo — `rigdoctor install [--check] [-y]` + GUI Setup tab.
+  The **user-local app install** is `install.sh` (private venv + `~/.local/bin` launchers +
+  desktop entry, no root; handles the `python3-venv` prerequisite) plus a self-extracting
+  **`.run`** (pure-Python self-extractor, `packaging/make_run.py`, built by CI). *Pending:*
+  config/module selection + `systemd --user`
+  service enable.
 - **M12 Session sharing / remote assist** (D16) — let a helper inspect a user's machine, in
   an escalating ladder: (1) **diagnostic bundle export** (inventory + recent log + report,
   one-way), (2) **live read-only view** over a user-chosen tunnel (Tailscale/cloudflared/SSH,
   no hosted relay), (3) **gated interactive terminal** wrapping tmate/sshx (read-only by
   default; read-write only on explicit consent — a deliberate exception to D9). Per-session
   consent, ephemeral revocable tokens, audit log.
-- **M13 Auto-update** (D18) — *planned.* On launch, check the public Gitea releases API and
+- **M13 Auto-update** (D18) — *check + auth implemented:* updates are **gated to Gitea account
+  holders** via a Personal Access Token, stored **encrypted in the OS keyring** (`secret-tool`)
+  with a 0600-file fallback (`config.load_token`/`save_token`/`token_backend`). `core/updates`
+  queries the releases API with the token; CLI `login`/`logout`/`update`; GUI Setup "Update
+  access" panel + sidebar states. The no-root **self-update apply** is implemented:
+  `rigdoctor update` runs an authenticated `pip install --upgrade "rigdoctor[gui] @
+  git+https://oauth2:<token>@…@<tag>"` into the user-local venv (GUI "Update to v…" button +
+  restart prompt; token scrubbed). Installed via the user-local **`install.sh`** /
+  self-extracting **`.run`** (M9).
+  *Original plan:* On launch, check the public Gitea releases API and
   **self-update a user-local install with no root** (download → verify checksum/signature →
   atomic symlink swap → restart, incl. the daemon). HTTPS-only, version-check-only (no
   telemetry), opt-out-able. Surfaced in the GUI; `rigdoctor update` in the CLI. (`.deb` users

docs/ROADMAP.md

@@ -15,8 +15,8 @@ Ubuntu + NVIDIA first; `.deb` distribution (see `DECISIONS.md`).
 - [x] M3 crash-capture logger (JSONL, fsync per sample, GPU-lost detection, size rotation)
 - [x] Manual trigger mode (`rigdoctor record run/start/stop/status`); `systemd --user`
       service + other trigger modes in Phase 4 (`run` is already the service entrypoint)
-- [ ] M4 health report (Xid/panic/OOM/MCE/AER/thermal scan + driver-mismatch + snapshot,
-      suggested fixes only — D9)
+- [x] M4 health report (Xid/panic/OOM/MCE/AER/thermal scan + SMART + driver-mismatch +
+      journald-persistence + live temps, suggested fixes only — D9; GPU-firmware verify deferred)
 - [x] `record report` post-crash summary (peak temps/power per subsystem, events, last N samples)
 - **Exit criteria:** user can run it during gaming and, after a freeze/black-screen, see the
   last readings + a plausible cause.
@@ -39,15 +39,18 @@ Ubuntu + NVIDIA first; `.deb` distribution (see `DECISIONS.md`).
 - [ ] Logger trigger modes: always-on + game-launch (D12 — wrapper first:
       `rigdoctor wrap %command%` + global Steam compat-tool; zero-config watcher
       (Steam RunningAppID + /proc) and GameMode hook follow)
-- [ ] M9 interactive installer (GPU detection, module menu, apt dependency resolution,
-      service enable + trigger-mode pick)
+- [~] M9 interactive installer — *done:* distro/GPU detection + optional-dependency install
+      (`rigdoctor install`, GUI Setup tab); **user-local `install.sh` + self-extracting `.run`**
+      (no-root venv install, handles python3-venv prereq, CI-built). *Pending:* module-selection
+      config + `systemd --user` service enable + trigger-mode pick.
 - [ ] `.deb` packaging (D8) declaring per-bundle deps incl. python3-pyside6 for Desktop UI
 
 ## Phase 5 — Breadth (later)
 - [ ] AMD GPU support in M1 (Steam Deck / Radeon)
 - [ ] Intel GPU best-effort
-- [ ] M13 auto-update (D18) — launch-time version check + no-root self-update of the
-      user-local install from the public Gitea releases; GUI prompt + `rigdoctor update`
+- [x] M13 auto-update (D18) — launch-time version check (GUI sidebar) + no-root self-update
+      apply (`rigdoctor update` / sidebar button → authenticated pip upgrade), token-gated.
+      Restart-after-update is manual for now.
 - [ ] (Later, separate milestone) Optional auto-apply of suggested fixes behind explicit
       consent — currently out of scope (D9)
 

install.sh

@@ -0,0 +1,117 @@
+#!/usr/bin/env sh
+# RigDoctor user-local installer (no root). Creates a private venv, links the
+# `rigdoctor` / `rigdoctor-gui` commands into ~/.local/bin, and adds a desktop
+# entry. Installs from a bundled wheel (the .run installer) or from a source
+# checkout. Re-run to upgrade; `./install.sh --uninstall` to remove.
+set -eu
+
+APP_NAME=rigdoctor
+DATA_HOME="${XDG_DATA_HOME:-$HOME/.local/share}"
+VENV="$DATA_HOME/$APP_NAME/venv"
+BIN_DIR="$HOME/.local/bin"
+DESKTOP_DIR="$DATA_HOME/applications"
+DESKTOP_FILE="$DESKTOP_DIR/rigdoctor.desktop"
+SCRIPT_DIR=$(CDPATH= cd -- "$(dirname -- "$0")" && pwd)
+
+uninstall() {
+    echo "Removing RigDoctor user-local install…"
+    rm -rf "$VENV"
+    rm -f "$BIN_DIR/rigdoctor" "$BIN_DIR/rigdoctor-gui" "$DESKTOP_FILE" \
+          "$DATA_HOME/icons/hicolor/scalable/apps/rigdoctor.svg"
+    echo "Done. (Config and logs under ~/.config/rigdoctor and ~/.local/share/rigdoctor were kept.)"
+}
+
+REF=""
+while [ $# -gt 0 ]; do
+    case "$1" in
+        --uninstall) uninstall; exit 0 ;;
+        --ref) REF="${2:-}"; [ -n "$REF" ] || { echo "--ref needs a tag"; exit 1; }; shift 2 ;;
+        -h|--help) echo "Usage: install.sh [--ref <tag>] [--uninstall]"; exit 0 ;;
+        *) echo "Unknown option: $1"; exit 1 ;;
+    esac
+done
+
+PY=python3
+command -v "$PY" >/dev/null 2>&1 || { echo "python3 not found — install Python 3.11+."; exit 1; }
+"$PY" - <<'EOF' || { echo "Python 3.11+ is required."; exit 1; }
+import sys
+sys.exit(0 if sys.version_info >= (3, 11) else 1)
+EOF
+
+# venv support (ensurepip) is required; install python3-venv if it's missing.
+if ! "$PY" -c "import ensurepip" >/dev/null 2>&1; then
+    PYVER=$("$PY" -c "import sys; print(f'{sys.version_info.major}.{sys.version_info.minor}')")
+    PKGS="python3-venv python${PYVER}-venv"
+    echo "Python venv support is missing — needs: $PKGS"
+    if command -v pkexec >/dev/null 2>&1; then ESC=pkexec
+    elif command -v sudo >/dev/null 2>&1; then ESC=sudo
+    else ESC=""; fi
+    if [ -n "$ESC" ] && command -v apt-get >/dev/null 2>&1; then
+        echo "Installing $PKGS (you may be prompted for your password)…"
+        "$ESC" sh -c "apt-get update && apt-get install -y $PKGS" \
+            || { echo "Failed. Install manually: sudo apt install $PKGS"; exit 1; }
+    else
+        echo "Install it manually, then re-run: sudo apt install $PKGS"
+        exit 1
+    fi
+fi
+
+# Where to install from: a specific released tag (--ref), a bundled wheel, or source.
+WHEEL=$(ls "$SCRIPT_DIR"/rigdoctor-*.whl 2>/dev/null | head -n1 || true)
+if [ -n "$REF" ]; then
+    CONF="${XDG_CONFIG_HOME:-$HOME/.config}/rigdoctor/token"
+    TOKEN="${RIGDOCTOR_TOKEN:-$(cat "$CONF" 2>/dev/null || true)}"
+    [ -n "$TOKEN" ] || { echo "--ref needs a token (run 'rigdoctor login' or set RIGDOCTOR_TOKEN)."; exit 1; }
+    SRC="rigdoctor[gui] @ git+https://oauth2:$TOKEN@git.jesseyvanofferen.com/jessey/rigdoctor.git@$REF"
+elif [ -n "$WHEEL" ]; then
+    SRC="$WHEEL[gui]"
+elif [ -f "$SCRIPT_DIR/pyproject.toml" ]; then
+    SRC="$SCRIPT_DIR[gui]"
+else
+    echo "No bundled wheel or source found next to the installer."
+    exit 1
+fi
+
+echo "Creating venv at $VENV…"
+"$PY" -m venv "$VENV"
+"$VENV/bin/pip" install --upgrade pip >/dev/null
+echo "Installing RigDoctor (pulls in PySide6 — this can take a minute)…"
+"$VENV/bin/pip" install "$SRC"
+
+mkdir -p "$BIN_DIR"
+ln -sf "$VENV/bin/rigdoctor" "$BIN_DIR/rigdoctor"
+ln -sf "$VENV/bin/rigdoctor-gui" "$BIN_DIR/rigdoctor-gui"
+
+# Install the app icon (for the dock/launcher); fall back to a stock icon.
+ICON_NAME=utilities-system-monitor
+ICON_SRC=$("$VENV/bin/python" -c "import os, rigdoctor.gui as g; print(os.path.join(os.path.dirname(g.__file__), 'assets', 'rigdoctor.svg'))" 2>/dev/null || true)
+if [ -n "$ICON_SRC" ] && [ -f "$ICON_SRC" ]; then
+    ICON_DST="$DATA_HOME/icons/hicolor/scalable/apps/rigdoctor.svg"
+    mkdir -p "$(dirname "$ICON_DST")"
+    cp "$ICON_SRC" "$ICON_DST"
+    ICON_NAME=rigdoctor
+    command -v gtk-update-icon-cache >/dev/null 2>&1 && gtk-update-icon-cache -qtf "$DATA_HOME/icons/hicolor" 2>/dev/null || true
+fi
+
+mkdir -p "$DESKTOP_DIR"
+cat > "$DESKTOP_FILE" <<EOF
+[Desktop Entry]
+Type=Application
+Name=RigDoctor
+Comment=Hardware monitoring & crash diagnostics for Linux gamers
+Exec=$VENV/bin/rigdoctor-gui
+Icon=$ICON_NAME
+Terminal=false
+Categories=System;Monitor;Utility;
+StartupWMClass=rigdoctor
+EOF
+command -v update-desktop-database >/dev/null 2>&1 && update-desktop-database "$DESKTOP_DIR" 2>/dev/null || true
+
+echo
+echo "RigDoctor $("$VENV/bin/rigdoctor" --version 2>/dev/null | awk '{print $2}') installed."
+echo "  GUI:  rigdoctor-gui   (or find 'RigDoctor' in your app menu)"
+echo "  CLI:  rigdoctor --help"
+case ":$PATH:" in
+    *":$BIN_DIR:"*) ;;
+    *) echo "  Note: add $BIN_DIR to your PATH (a fresh login usually does this).";;
+esac

packaging/bump.sh

@@ -0,0 +1,25 @@
+#!/usr/bin/env sh
+# Auto-set the next version from Conventional Commits (git-cliff), per D21.
+# Run after committing your feat:/fix: changes; it updates __init__.py + pyproject.toml.
+# Then update CHANGELOG.md, commit as `chore(release): vX.Y.Z`, and push (CI tags + releases).
+set -eu
+
+ROOT=$(CDPATH= cd -- "$(dirname -- "$0")/.." && pwd)
+cd "$ROOT"
+
+command -v git-cliff >/dev/null 2>&1 || { echo "git-cliff not found. Install: pip install git-cliff"; exit 1; }
+
+NEXT=$(git-cliff --bumped-version | sed 's/^v//')
+[ -n "$NEXT" ] || { echo "Could not compute the next version."; exit 1; }
+
+python3 - "$NEXT" <<'PY'
+import pathlib, re, sys
+version = sys.argv[1]
+init = pathlib.Path("src/rigdoctor/__init__.py")
+init.write_text(re.sub(r'__version__ = "[^"]+"', f'__version__ = "{version}"', init.read_text()))
+proj = pathlib.Path("pyproject.toml")
+proj.write_text(re.sub(r'(?m)^version = "[^"]+"', f'version = "{version}"', proj.read_text(), count=1))
+PY
+
+echo "Set version to $NEXT."
+echo "Next: add a '## [$NEXT]' CHANGELOG section, then commit as 'chore(release): v$NEXT'."

packaging/changelog.sh

@@ -0,0 +1,20 @@
+#!/usr/bin/env sh
+# Regenerate CHANGELOG.md from Conventional Commits using git-cliff (D20).
+# Install once:  pip install git-cliff   (ships prebuilt binaries)
+# Usage:         packaging/changelog.sh [--tag vX.Y.Z]
+set -eu
+
+ROOT=$(CDPATH= cd -- "$(dirname -- "$0")/.." && pwd)
+cd "$ROOT"
+
+command -v git-cliff >/dev/null 2>&1 || {
+    echo "git-cliff not found. Install it: pip install git-cliff"
+    exit 1
+}
+
+if [ "${1:-}" = "--tag" ] && [ -n "${2:-}" ]; then
+    git-cliff --tag "$2" -o CHANGELOG.md
+else
+    git-cliff -o CHANGELOG.md
+fi
+echo "Wrote CHANGELOG.md"

packaging/make-run.sh

@@ -0,0 +1,3 @@
+#!/usr/bin/env sh
+# Build the self-extracting .run installer (delegates to make_run.py — no makeself).
+exec python3 "$(CDPATH= cd -- "$(dirname -- "$0")" && pwd)/make_run.py" "$@"

packaging/make_run.py

@@ -0,0 +1,62 @@
+#!/usr/bin/env python3
+"""Build a dependency-free self-extracting .run installer (no makeself).
+
+Produces dist/rigdoctor-<version>-installer.run: a POSIX shell stub with an appended
+tar.gz of the wheel + install.sh. Running it extracts to a temp dir and runs install.sh.
+"""
+
+from __future__ import annotations
+
+import io
+import os
+import subprocess
+import sys
+import tarfile
+import tomllib
+from pathlib import Path
+
+ROOT = Path(__file__).resolve().parent.parent
+MARKER = "__RIGDOCTOR_ARCHIVE__"
+
+STUB = f"""#!/bin/sh
+# RigDoctor self-extracting installer. Extracts the embedded archive and runs install.sh.
+set -eu
+SKIP=$(awk '/^{MARKER}$/ {{ print NR + 1; exit 0 }}' "$0")
+TMP=$(mktemp -d)
+tail -n +"$SKIP" "$0" | tar -xz -C "$TMP"
+sh "$TMP/install.sh" "$@"
+RET=$?
+rm -rf "$TMP"
+exit $RET
+{MARKER}
+"""
+
+
+def main() -> int:
+    version = tomllib.loads((ROOT / "pyproject.toml").read_text())["project"]["version"]
+    dist = ROOT / "dist"
+    dist.mkdir(exist_ok=True)
+
+    wheel = dist / f"rigdoctor-{version}-py3-none-any.whl"
+    if not wheel.exists():
+        subprocess.run([sys.executable, "-m", "build", "--wheel"], cwd=ROOT, check=True)
+    if not wheel.exists():
+        print(f"wheel not found: {wheel}", file=sys.stderr)
+        return 1
+
+    buf = io.BytesIO()
+    with tarfile.open(fileobj=buf, mode="w:gz") as tar:
+        tar.add(wheel, arcname=wheel.name)
+        tar.add(ROOT / "install.sh", arcname="install.sh")
+
+    out = dist / f"rigdoctor-{version}-installer.run"
+    with open(out, "wb") as f:
+        f.write(STUB.encode())
+        f.write(buf.getvalue())
+    os.chmod(out, 0o755)
+    print(f"Built {out}")
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "rigdoctor"
-version = "0.0.2"
+version = "0.4.0"
 description = "Modular hardware monitoring & crash diagnostics for Linux gamers."
 readme = "README.md"
 requires-python = ">=3.11"
@@ -21,3 +21,6 @@ rigdoctor-gui = "rigdoctor.gui.app:main"
 
 [tool.setuptools.packages.find]
 where = ["src"]
+
+[tool.setuptools.package-data]
+rigdoctor = ["gui/assets/*.svg"]

src/rigdoctor/__init__.py

@@ -1,3 +1,3 @@
 """RigDoctor — modular hardware monitoring & crash diagnostics for Linux gamers."""
 
-__version__ = "0.0.2"
+__version__ = "0.4.0"

src/rigdoctor/cli.py

@@ -164,9 +164,179 @@ def cmd_record_report(args) -> int:
     return 0
 
 
+def cmd_install(args) -> int:
+    from .core import installer, sysenv
+
+    print(f"Distro:          {sysenv.distro_name()}")
+    pm = sysenv.package_manager()
+    print(f"Package manager: {pm or 'none (only apt is supported)'}")
+    print(f"GPU:             {', '.join(sysenv.gpu_vendors()) or 'unknown'}\n")
+
+    status = installer.component_status()
+    print("Optional components:")
+    for component, present in status:
+        mark = "✓" if present else "✗"
+        print(f"  [{mark}] {component.name:<22} — {component.enables}")
+        if not present:
+            print(f"        apt: {' '.join(component.apt)}")
+
+    missing = [c for c, present in status if not present]
+    if not missing:
+        print("\nAll optional components are installed. ✔")
+        return 0
+
+    packages = installer.missing_packages(missing)
+    print(f"\nMissing packages: {' '.join(packages)}")
+    if args.check:
+        return 0
+    if pm != "apt":
+        print(f"Automatic install needs apt. Install manually:\n  sudo apt install {' '.join(packages)}")
+        return 1
+    if not args.yes:
+        try:
+            reply = input(f"\nInstall {len(packages)} package(s) now? [y/N] ").strip().lower()
+        except EOFError:
+            reply = "n"
+        if reply not in ("y", "yes"):
+            print("Aborted.")
+            return 1
+
+    print("Installing (you may be prompted for your password)…")
+    rc, out = installer.install_packages(packages)
+    print(out[-2000:])
+    if rc == 0:
+        still = [c.name for c, present in installer.component_status() if not present]
+        print("\nStill missing: " + (", ".join(still) if still else "none ✔"))
+    else:
+        print(f"\nInstall failed (exit {rc}).")
+    return rc
+
+
+def cmd_login(args) -> int:
+    from getpass import getpass
+
+    from .core import updates
+
+    token = args.token
+    if not token:
+        print(f"Create a token (scope read:repository) at: {updates.TOKEN_PAGE}")
+        try:
+            token = getpass("Paste token: ").strip()
+        except (EOFError, KeyboardInterrupt):
+            token = ""
+    if not token:
+        print("No token provided.")
+        return 1
+    config.save_token(token)
+    state, tag, _notes = updates.update_state()
+    if state == updates.AUTH:
+        print("Token saved, but the server rejected it (check scope/permissions).")
+        return 1
+    if state in (updates.UP_TO_DATE, updates.AVAILABLE):
+        print(f"Token saved and verified. Latest release: {tag}.")
+        return 0
+    print("Token saved (couldn't reach the server to verify right now).")
+    return 0
+
+
+def cmd_logout(args) -> int:
+    config.clear_token()
+    print("Update token removed.")
+    return 0
+
+
+def cmd_update(args) -> int:
+    from .core import updates
+
+    state, tag, notes = updates.update_state()
+    if state == updates.NO_TOKEN:
+        print("No update token. Run `rigdoctor login` after creating one at:")
+        print(f"  {updates.TOKEN_PAGE}")
+        return 1
+    if state == updates.AUTH:
+        print("The update server rejected your token (check scope/permissions).")
+        return 1
+    if state == updates.NETWORK:
+        print("Couldn't reach the update server.")
+        return 1
+    if state == updates.UP_TO_DATE:
+        print(f"Up to date (v{__version__}).")
+        return 0
+    # AVAILABLE
+    print(f"Update available: {tag}  (current v{__version__}).")
+    if notes:
+        print("\nWhat's new:\n" + "\n".join("  " + ln for ln in notes.splitlines()) + "\n")
+    if args.check:
+        return 0
+    print(f"Installing {tag}…")
+    rc, out = updates.apply_update(tag)
+    print(out[-2000:])
+    if rc == 0:
+        print(f"\nUpdated to {tag}. Restart RigDoctor to use the new version.")
+        return 0
+    print(f"\nUpdate failed (exit {rc}).")
+    return rc
+
+
+def cmd_uninstall(args) -> int:
+    from .core import uninstall as uninstaller
+
+    scope = "everything (app + settings, token, and logs)" if args.purge else "the app (settings/logs kept)"
+    if not args.yes:
+        try:
+            reply = input(f"Uninstall RigDoctor — remove {scope}? [y/N] ").strip().lower()
+        except EOFError:
+            reply = "n"
+        if reply not in ("y", "yes"):
+            print("Aborted.")
+            return 1
+    uninstaller.uninstall(purge=args.purge)
+    print("Uninstalling… RigDoctor will be removed momentarily.")
+    return 0
+
+
+def cmd_collect_priv(args) -> int:
+    """Internal: emit root-only data (SMART + dmidecode) as JSON, run via pkexec at launch."""
+    from dataclasses import asdict
+
+    from .core.health import check_smart
+    from .core.inventory import _dmidecode
+
+    data = {"smart": [asdict(f) for f in check_smart()], "dmidecode": _dmidecode()}
+    print(json.dumps(data))
+    return 0
+
+
+def cmd_inventory(args) -> int:
+    from .core import inventory
+
+    sections = inventory.collect()
+    if args.json:
+        text = inventory.render_json(sections)
+    elif args.markdown:
+        text = inventory.render_markdown(sections)
+    else:
+        text = inventory.render_text(sections)
+    if args.output:
+        Path(args.output).write_text(text)
+        print(f"Wrote {args.output}")
+    else:
+        print(text)
+    return 0
+
+
 def cmd_report(args) -> int:
-    print("`report` (M4 health report) is not implemented yet — next on the roadmap.")
-    return 2
+    from dataclasses import asdict
+
+    from .core.health import run_health_checks
+    from .render import render_health
+
+    findings = run_health_checks()
+    if args.json:
+        print(json.dumps([asdict(f) for f in findings], indent=2, ensure_ascii=False))
+    else:
+        print(render_health(findings))
+    return 0
 
 
 def build_parser() -> argparse.ArgumentParser:
@@ -188,6 +358,25 @@ def build_parser() -> argparse.ArgumentParser:
     sub.add_parser("gui", help="launch the desktop GUI (needs PySide6)").set_defaults(func=cmd_gui)
     sub.add_parser("sources", help="list detected sensor sources").set_defaults(func=cmd_sources)
 
+    inst = sub.add_parser("install", help="set up optional system dependencies (M9)")
+    inst.add_argument("--check", action="store_true", help="report status only; install nothing")
+    inst.add_argument("-y", "--yes", action="store_true", help="install without confirmation")
+    inst.set_defaults(func=cmd_install)
+
+    login = sub.add_parser("login", help="save a Gitea token for updates (M13)")
+    login.add_argument("--token", default=None, help="token (prompted if omitted)")
+    login.set_defaults(func=cmd_login)
+    sub.add_parser("logout", help="remove the saved update token").set_defaults(func=cmd_logout)
+
+    upd = sub.add_parser("update", help="check for / apply a newer version (M13)")
+    upd.add_argument("--check", action="store_true", help="only report, don't apply")
+    upd.set_defaults(func=cmd_update)
+
+    unin = sub.add_parser("uninstall", help="remove the user-local install")
+    unin.add_argument("--purge", action="store_true", help="also remove settings, token, and logs")
+    unin.add_argument("-y", "--yes", action="store_true", help="don't ask for confirmation")
+    unin.set_defaults(func=cmd_uninstall)
+
     rec = sub.add_parser("record", help="crash-capture logger (M3)")
     rec_sub = rec.add_subparsers(dest="record_cmd", required=True)
 
@@ -209,7 +398,18 @@ def build_parser() -> argparse.ArgumentParser:
     report_p.add_argument("--log", default=None, help="path to a capture log")
     report_p.set_defaults(func=cmd_record_report)
 
-    sub.add_parser("report", help="health report (coming soon)").set_defaults(func=cmd_report)
+    rep = sub.add_parser("report", help="health report (M4): scan logs/SMART/driver for issues")
+    rep.add_argument("--json", action="store_true", help="output JSON instead of text")
+    rep.set_defaults(func=cmd_report)
+
+    cp = sub.add_parser("collect-priv", help=argparse.SUPPRESS)  # internal: run via pkexec
+    cp.set_defaults(func=cmd_collect_priv)
+
+    inv = sub.add_parser("inventory", help="system inventory (M5): export hardware/OS details")
+    inv.add_argument("--json", action="store_true", help="output JSON")
+    inv.add_argument("--markdown", action="store_true", help="output Markdown (for forum/bug reports)")
+    inv.add_argument("-o", "--output", default=None, help="write to a file instead of stdout")
+    inv.set_defaults(func=cmd_inventory)
     return p
 
 

src/rigdoctor/config.py

@@ -3,6 +3,8 @@
 from __future__ import annotations
 
 import os
+import shutil
+import subprocess
 from pathlib import Path
 
 APP = "rigdoctor"
@@ -25,10 +27,121 @@ STATUS_FILE = STATE_DIR / "recorder.json"
 PID_FILE = STATE_DIR / "recorder.pid"
 SPAWN_LOG = STATE_DIR / "recorder.out"
 
+# Update access token (M13) — gates updates to Gitea account holders (D18).
+# Stored in the OS keyring (Secret Service / GNOME Keyring) via `secret-tool` when
+# available — encrypted at rest, unlocked with the login session — else a 0600 file.
+TOKEN_FILE = CONFIG_DIR / "token"
+_SECRET_ATTRS = ["application", "rigdoctor", "type", "update-token"]
+
+
+def _secret_tool() -> str | None:
+    return shutil.which("secret-tool")
+
+
+def keyring_available() -> bool:
+    """True if an encrypted OS keyring (secret-tool) is usable."""
+    return _secret_tool() is not None
+
+
+def _keyring_store(token: str) -> bool:
+    tool = _secret_tool()
+    if not tool:
+        return False
+    try:
+        proc = subprocess.run(
+            [tool, "store", "--label", "RigDoctor update token", *_SECRET_ATTRS],
+            input=token, text=True, capture_output=True, timeout=20,
+        )
+        return proc.returncode == 0
+    except (subprocess.SubprocessError, OSError):
+        return False
+
+
+def _keyring_lookup() -> str | None:
+    tool = _secret_tool()
+    if not tool:
+        return None
+    try:
+        proc = subprocess.run(
+            [tool, "lookup", *_SECRET_ATTRS], text=True, capture_output=True, timeout=20
+        )
+        if proc.returncode == 0 and proc.stdout.strip():
+            return proc.stdout.strip()
+    except (subprocess.SubprocessError, OSError):
+        pass
+    return None
+
+
+def _keyring_clear() -> None:
+    tool = _secret_tool()
+    if not tool:
+        return
+    try:
+        subprocess.run([tool, "clear", *_SECRET_ATTRS], capture_output=True, timeout=20)
+    except (subprocess.SubprocessError, OSError):
+        pass
+
+
+def load_token() -> str | None:
+    """Token from $RIGDOCTOR_TOKEN, then the OS keyring, then a 0600 file."""
+    env = os.environ.get("RIGDOCTOR_TOKEN")
+    if env and env.strip():
+        return env.strip()
+    from_keyring = _keyring_lookup()
+    if from_keyring:
+        return from_keyring
+    try:
+        token = TOKEN_FILE.read_text().strip()
+        return token or None
+    except OSError:
+        return None
+
+
+def save_token(token: str) -> None:
+    """Save to the OS keyring if possible (encrypted); else a 0600 file."""
+    token = token.strip()
+    if _keyring_store(token):
+        try:  # don't leave a plaintext copy once it's in the keyring
+            TOKEN_FILE.unlink()
+        except OSError:
+            pass
+        return
+    CONFIG_DIR.mkdir(parents=True, exist_ok=True)
+    TOKEN_FILE.write_text(token + "\n")
+    try:
+        TOKEN_FILE.chmod(0o600)
+    except OSError:
+        pass
+
+
+def clear_token() -> None:
+    _keyring_clear()
+    try:
+        TOKEN_FILE.unlink()
+    except OSError:
+        pass
+
+
+def token_backend() -> str:
+    """Where the active token lives: 'env' | 'keyring' | 'file' | 'none'."""
+    env = os.environ.get("RIGDOCTOR_TOKEN")
+    if env and env.strip():
+        return "env"
+    if _keyring_lookup() is not None:
+        return "keyring"
+    if TOKEN_FILE.exists():
+        return "file"
+    return "none"
+
 DEFAULTS: dict = {
     "interval": 1.0,           # sampling interval in seconds (default ≤1 Hz — NFR)
     "log_max_bytes": 20_000_000,  # rotate a log segment past this size
     "log_backups": 10,            # keep this many rotated segments (bounds disk use)
+    "update_check_minutes": 30,   # re-check for updates this often while running (0 = off)
+    "elevate_on_launch": True,    # GUI asks for the password once at launch (SMART/dmidecode)
+    "alerts_enabled": True,       # desktop notifications on overheat / GPU-lost / new version
+    "gpu_temp_alert": 90.0,       # °C — alert when GPU reaches this
+    "cpu_temp_alert": 95.0,       # °C — alert when CPU reaches this
 }
 
 
@@ -44,3 +157,27 @@ def load_config() -> dict:
     except Exception:
         pass
     return cfg
+
+
+def _toml_value(value) -> str:
+    if isinstance(value, bool):
+        return "true" if value else "false"
+    if isinstance(value, (int, float)):
+        return repr(value)
+    return '"' + str(value).replace("\\", "\\\\").replace('"', '\\"') + '"'
+
+
+def save_config(values: dict) -> None:
+    """Write a flat config.toml (stdlib has no TOML writer)."""
+    CONFIG_DIR.mkdir(parents=True, exist_ok=True)
+    lines = ["# RigDoctor config — edit in the app (Notifications) or here."]
+    lines += [f"{key} = {_toml_value(value)}" for key, value in values.items()]
+    CONFIG_FILE.write_text("\n".join(lines) + "\n")
+
+
+def update_config(**changes) -> dict:
+    """Merge changes into the current effective config and persist them."""
+    cfg = load_config()
+    cfg.update(changes)
+    save_config(cfg)
+    return cfg

src/rigdoctor/core/alerts.py

@@ -0,0 +1,91 @@
+"""Desktop alerts (M8): notify on overheat / GPU-lost / new version via notify-send.
+
+Edge-triggered: an alert fires when a condition becomes true (not every sample), and
+can fire again only after it has cleared and a cooldown has passed — so a hot GPU or a
+1-Hz sample loop doesn't spam notifications. Degrades to a no-op if notify-send is absent.
+"""
+
+from __future__ import annotations
+
+import shutil
+import subprocess
+import time
+
+from .sample import Sample
+
+APP_NAME = "RigDoctor"
+_ICON = "utilities-system-monitor"
+
+
+def available() -> bool:
+    return shutil.which("notify-send") is not None
+
+
+def notify(title: str, message: str, urgency: str = "normal") -> bool:
+    """Send a desktop notification (best-effort). urgency: low|normal|critical."""
+    if not available():
+        return False
+    try:
+        subprocess.run(
+            ["notify-send", "-a", APP_NAME, "-u", urgency, "-i", _ICON, title, message],
+            timeout=10,
+            check=False,
+        )
+        return True
+    except (subprocess.SubprocessError, OSError):
+        return False
+
+
+class AlertMonitor:
+    """Evaluate samples and raise edge-triggered desktop alerts."""
+
+    def __init__(self, gpu_temp: float = 90.0, cpu_temp: float = 95.0, cooldown: float = 300.0):
+        self.gpu_temp = gpu_temp
+        self.cpu_temp = cpu_temp
+        self.cooldown = cooldown
+        self.enabled = True
+        self._active: dict[str, bool] = {}
+        self._last: dict[str, float] = {}
+
+    def _fire(self, key: str, title: str, message: str, urgency: str = "critical") -> None:
+        if self._active.get(key):
+            return  # already alerting; wait until it clears
+        now = time.time()
+        if now - self._last.get(key, 0.0) < self.cooldown:
+            return
+        self._active[key] = True
+        self._last[key] = now
+        notify(title, message, urgency)
+
+    def _clear(self, key: str) -> None:
+        self._active[key] = False
+
+    def check(self, sample: Sample) -> None:
+        if not self.enabled:
+            return
+        gpu_t = next(
+            (r.value for r in sample.readings
+             if r.source == "gpu" and r.metric == "temp" and r.label == "" and r.value is not None),
+            None,
+        )
+        if gpu_t is not None:
+            if gpu_t >= self.gpu_temp:
+                self._fire("gpu_temp", "GPU overheating", f"GPU at {gpu_t:.0f} °C")
+            else:
+                self._clear("gpu_temp")
+
+        cpu_temps = [r.value for r in sample.readings
+                     if r.source == "cpu" and r.metric == "temp" and r.value is not None]
+        if cpu_temps:
+            cpu_t = max(cpu_temps)
+            if cpu_t >= self.cpu_temp:
+                self._fire("cpu_temp", "CPU overheating", f"CPU at {cpu_t:.0f} °C")
+            else:
+                self._clear("cpu_temp")
+
+        lost = any(r.source == "gpu" and r.metric == "status" and r.label == "query-timeout"
+                   for r in sample.readings)
+        if lost:
+            self._fire("gpu_lost", "GPU not responding", "nvidia-smi query timed out — the GPU may have dropped")
+        else:
+            self._clear("gpu_lost")

src/rigdoctor/core/catalog.py

@@ -0,0 +1,48 @@
+"""Installable component catalog (M9): optional system tools and what they enable.
+
+apt-only (D15). Core monitoring (M1/M3/M4) needs no packages — these are optional
+enrichments the installer can add. Each component is detected by a representative
+command (present == usable).
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+
+
+@dataclass(frozen=True)
+class Component:
+    id: str
+    name: str
+    bundle: str
+    enables: str          # capability unlocked when present
+    apt: tuple[str, ...]  # apt package name(s)
+    command: str          # command used to detect presence
+
+
+COMPONENTS: tuple[Component, ...] = (
+    Component(
+        "smartmontools", "SMART disk health", "Diagnostics",
+        "Disk health (SMART) in the health report", ("smartmontools",), "smartctl",
+    ),
+    Component(
+        "lm-sensors", "lm-sensors", "Diagnostics",
+        "Extra motherboard / voltage sensors", ("lm-sensors",), "sensors",
+    ),
+    Component(
+        "dmidecode", "dmidecode", "Diagnostics",
+        "Motherboard / BIOS / RAM details for system inventory", ("dmidecode",), "dmidecode",
+    ),
+    Component(
+        "pciutils", "pciutils", "Diagnostics",
+        "PCIe topology + GPU detection (lspci)", ("pciutils",), "lspci",
+    ),
+    Component(
+        "libnotify", "Desktop notifications", "Monitoring",
+        "Desktop alert notifications", ("libnotify-bin",), "notify-send",
+    ),
+    Component(
+        "libsecret", "Encrypted token storage", "Updates",
+        "Store the update token in the OS keyring, encrypted", ("libsecret-tools",), "secret-tool",
+    ),
+)

src/rigdoctor/core/elevation.py

@@ -0,0 +1,51 @@
+"""Session privilege elevation.
+
+At GUI launch the app asks for the password once (pkexec) and collects the data that
+needs root — SMART health + dmidecode (board/BIOS/RAM) — caching it for the session so
+Health and Inventory can always show the full picture without per-action prompts.
+"""
+
+from __future__ import annotations
+
+import json
+import os
+import shutil
+import subprocess
+import sys
+
+_privileged: dict | None = None
+
+
+def privileged() -> dict | None:
+    """Cached root-collected data ({"smart": [...], "dmidecode": {...}}), or None."""
+    return _privileged
+
+
+def set_privileged(data: dict | None) -> None:
+    global _privileged
+    _privileged = data
+
+
+def available() -> bool:
+    return shutil.which("pkexec") is not None and os.geteuid() != 0
+
+
+def _cli() -> list[str]:
+    candidate = os.path.join(os.path.dirname(sys.executable), "rigdoctor")
+    return [candidate] if os.path.exists(candidate) else [sys.executable, "-m", "rigdoctor"]
+
+
+def collect_via_pkexec(timeout: float = 120.0) -> dict | None:
+    """Run one elevated collection (single password prompt). None if unavailable/cancelled."""
+    if not available():
+        return None
+    try:
+        proc = subprocess.run(
+            ["pkexec", *_cli(), "collect-priv"],
+            capture_output=True, text=True, timeout=timeout,
+        )
+        if proc.returncode == 0 and proc.stdout.strip():
+            return json.loads(proc.stdout)
+    except (subprocess.SubprocessError, OSError, ValueError):
+        pass
+    return None

src/rigdoctor/core/health.py

@@ -0,0 +1,255 @@
+"""Health report (M4): scan kernel logs + SMART + driver/library state into a
+prioritized, plain-language findings list with suggested fixes (read-only, D9).
+
+Stdlib-only. Every check degrades gracefully — a missing tool/permission yields an
+info finding, never an exception.
+"""
+
+from __future__ import annotations
+
+import re
+import shutil
+import subprocess
+from dataclasses import dataclass
+from pathlib import Path
+
+CRITICAL = "critical"
+WARNING = "warning"
+INFO = "info"
+OK = "ok"
+_ORDER = {CRITICAL: 0, WARNING: 1, INFO: 2, OK: 3}
+
+
+@dataclass
+class Finding:
+    severity: str   # critical | warning | info | ok
+    category: str   # GPU, Kernel, Memory, Storage, Thermal, Driver, PCIe, Logs
+    title: str
+    detail: str = ""
+    suggestion: str = ""
+
+
+# --- NVIDIA Xid knowledge (the seed crash is Xid 79) --------------------------
+_XID_INFO: dict[int, tuple[str, str]] = {
+    13: (WARNING, "Graphics engine exception (often an app/driver bug or unstable overclock)"),
+    31: (WARNING, "GPU memory page fault (usually a driver or application bug)"),
+    43: (WARNING, "GPU stopped processing a task (application error)"),
+    45: (INFO, "Preemptive channel removal (often a side-effect of another error or a reboot)"),
+    48: (CRITICAL, "Double-bit ECC error — VRAM hardware fault"),
+    62: (CRITICAL, "Internal microcontroller halt (often follows instability)"),
+    79: (CRITICAL, "GPU has fallen off the bus — hardware: power delivery, PCIe link, or thermals"),
+    94: (CRITICAL, "Contained ECC error"),
+    95: (CRITICAL, "Uncontained ECC error"),
+    119: (CRITICAL, "GSP RPC timeout — GPU System Processor hang"),
+    120: (CRITICAL, "GSP error — GPU System Processor fault"),
+}
+_XID_SUGGEST: dict[int, str] = {
+    79: "Check PSU/power cables and reseat the GPU/riser; test a lower power limit "
+        "(`sudo nvidia-smi -pl <watts>`) and capture a session with `rigdoctor record`.",
+    48: "Persistent VRAM ECC errors mean failing memory — RMA the card if it recurs.",
+    119: "GSP hangs are often driver-version specific — try a different driver branch.",
+    120: "GSP errors are often driver-version specific — try a different driver branch.",
+}
+_XID_RE = re.compile(r"Xid(?:\s*\([^)]*\))?:?\s*(\d+)")
+
+
+def scan_journal_text(text: str) -> list[Finding]:
+    """Parse kernel-log text into findings (separated from IO so it's testable)."""
+    lines = text.splitlines()
+    findings: list[Finding] = []
+
+    xids: dict[int, int] = {}
+    for line in lines:
+        if "Xid" in line:
+            m = _XID_RE.search(line)
+            if m:
+                code = int(m.group(1))
+                xids[code] = xids.get(code, 0) + 1
+    for code in sorted(xids):
+        severity, desc = _XID_INFO.get(code, (WARNING, f"NVIDIA GPU error (Xid {code})"))
+        suggest = _XID_SUGGEST.get(code, "Look up this Xid code in NVIDIA's Xid error documentation.")
+        findings.append(Finding(severity, "GPU", f"NVIDIA Xid {code} ×{xids[code]}", desc, suggest))
+
+    oom = sum(1 for ln in lines if "Out of memory" in ln or "oom-kill" in ln or "oom_reaper" in ln)
+    if oom:
+        findings.append(Finding(
+            WARNING, "Memory", f"Out-of-memory kills ×{oom}",
+            "The kernel killed processes to reclaim RAM.",
+            "Close memory-heavy apps, add zram/swap, or investigate a leak.",
+        ))
+
+    if any("Kernel panic" in ln for ln in lines):
+        findings.append(Finding(
+            CRITICAL, "Kernel", "Kernel panic recorded",
+            "The kernel hit an unrecoverable error.",
+            "Note the panic message; review recent driver/kernel updates and hardware.",
+        ))
+
+    if any("mce:" in ln or "Machine check" in ln or "Hardware Error" in ln for ln in lines):
+        findings.append(Finding(
+            CRITICAL, "Hardware", "Machine Check Exception (MCE)",
+            "The CPU reported a hardware error.",
+            "Run memtest86 for RAM, check CPU temps/voltages, and review the MCE detail.",
+        ))
+
+    if any("AER:" in ln or "PCIe Bus Error" in ln or ("pcieport" in ln and "error" in ln.lower()) for ln in lines):
+        findings.append(Finding(
+            WARNING, "PCIe", "PCIe bus errors (AER)",
+            "Correctable/uncorrectable PCIe errors were logged.",
+            "Reseat the device and check risers/cabling; AER storms can precede a GPU drop.",
+        ))
+
+    low = [ln.lower() for ln in lines]
+    if any(("thermal" in ln and ("critical" in ln or "throttl" in ln)) or "temperature above threshold" in ln for ln in low):
+        findings.append(Finding(
+            WARNING, "Thermal", "Thermal events logged",
+            "The system logged thermal throttling / critical-temperature events.",
+            "Improve airflow/cooling and check fan curves; watch live temps on the dashboard.",
+        ))
+
+    if any("amdgpu" in ln and "reset" in ln for ln in low):
+        findings.append(Finding(
+            CRITICAL, "GPU", "AMD GPU reset (amdgpu)",
+            "The AMD GPU was reset after a hang.",
+            "Check power/thermals/driver; capture a session with `rigdoctor record`.",
+        ))
+
+    return findings
+
+
+def _journalctl(args: list[str]) -> str | None:
+    if shutil.which("journalctl") is None:
+        return None
+    try:
+        proc = subprocess.run(["journalctl", *args], capture_output=True, text=True, timeout=25)
+        return proc.stdout
+    except (subprocess.SubprocessError, OSError):
+        return None
+
+
+def check_journal() -> list[Finding]:
+    out = _journalctl(["-k", "--no-pager", "-o", "cat", "--since", "-7 days"])
+    if out is None:
+        return [Finding(
+            INFO, "Logs", "Couldn't read the kernel journal",
+            "journalctl is unavailable or not readable.",
+            "Ensure systemd/journald is present and your user is in the 'systemd-journal' or 'adm' group.",
+        )]
+    findings = scan_journal_text(out)
+    if not findings:
+        findings.append(Finding(
+            OK, "Logs", "No notable kernel errors (last 7 days)",
+            "No Xid, panic, OOM, MCE, PCIe AER, or thermal events found.",
+        ))
+    return findings
+
+
+def check_journal_persistence() -> list[Finding]:
+    if Path("/var/log/journal").is_dir():
+        return []
+    return [Finding(
+        WARNING, "Logs", "journald isn't persistent across reboots",
+        "Crash-boot kernel logs are discarded on reboot, so a hard freeze's evidence can vanish.",
+        "Enable persistent logging: `sudo mkdir -p /var/log/journal && sudo systemctl restart systemd-journald`",
+    )]
+
+
+def check_nvidia_driver() -> list[Finding]:
+    if shutil.which("nvidia-smi") is None:
+        return []
+    try:
+        proc = subprocess.run(["nvidia-smi"], capture_output=True, text=True, timeout=10)
+    except (subprocess.SubprocessError, OSError):
+        return []
+    if "Driver/library version mismatch" in (proc.stdout + proc.stderr):
+        return [Finding(
+            CRITICAL, "Driver", "NVIDIA driver/library version mismatch",
+            "The loaded kernel module and the userspace NVIDIA libraries differ — GPU monitoring will fail until resolved.",
+            "Reboot to load the matching module (or finish the interrupted driver update).",
+        )]
+    return []
+
+
+def _smart_devices() -> list[str]:
+    try:
+        proc = subprocess.run(["smartctl", "--scan"], capture_output=True, text=True, timeout=10)
+    except (subprocess.SubprocessError, OSError):
+        return []
+    devices = []
+    for line in proc.stdout.splitlines():
+        line = line.strip()
+        if line.startswith("/dev/"):
+            devices.append(line.split()[0])
+    return devices
+
+
+def check_smart() -> list[Finding]:
+    if shutil.which("smartctl") is None:
+        return [Finding(
+            INFO, "Storage", "SMART not checked (smartmontools missing)",
+            "Disk self-health couldn't be read.",
+            "Install it for disk health checks: `sudo apt install smartmontools`",
+        )]
+    devices = _smart_devices()
+    if not devices:
+        return [Finding(
+            INFO, "Storage", "SMART: couldn't enumerate drives",
+            "Reading SMART usually needs root.",
+            "Run: `sudo rigdoctor report`",
+        )]
+    findings: list[Finding] = []
+    for dev in devices:
+        try:
+            proc = subprocess.run(["smartctl", "-H", dev], capture_output=True, text=True, timeout=15)
+        except (subprocess.SubprocessError, OSError):
+            continue
+        combined = proc.stdout + proc.stderr
+        if "Permission denied" in combined or "requires root" in combined.lower():
+            findings.append(Finding(INFO, "Storage", f"SMART for {dev} needs root", "", "Run: `sudo rigdoctor report`"))
+        elif "PASSED" in combined:
+            findings.append(Finding(OK, "Storage", f"SMART OK: {dev}", "Overall-health self-assessment passed."))
+        elif "FAILED" in combined or "FAILING_NOW" in combined:
+            findings.append(Finding(CRITICAL, "Storage", f"SMART FAILED: {dev}", "The drive reports failing health.", "Back up now and replace the drive."))
+    return findings
+
+
+def check_live_temps() -> list[Finding]:
+    from .sampler import Sampler
+    from .sources import available_sources
+
+    sample = Sampler(available_sources()).sample()
+    hot = [
+        (r.source, r.label or r.metric, r.value)
+        for r in sample.readings
+        if r.unit == "°C" and r.value is not None and r.value >= 90
+    ]
+    if not hot:
+        return []
+    worst = max(hot, key=lambda x: x[2])
+    detail = "; ".join(f"{s} {label} {v:.0f}°C" for s, label, v in hot)
+    return [Finding(
+        WARNING, "Thermal", f"High temperature right now ({worst[2]:.0f}°C)",
+        detail, "Check cooling/airflow and reduce load.",
+    )]
+
+
+def run_health_checks() -> list[Finding]:
+    """Run all checks and return findings sorted by severity (worst first).
+
+    SMART needs root; if the session collected it via launch elevation, use that
+    instead of re-running smartctl (which would just report "needs root").
+    """
+    from . import elevation
+
+    findings: list[Finding] = []
+    findings += check_nvidia_driver()
+    findings += check_journal()
+    findings += check_journal_persistence()
+    priv = elevation.privileged()
+    if priv is not None and priv.get("smart") is not None:
+        findings += [Finding(**d) for d in priv["smart"]]
+    else:
+        findings += check_smart()
+    findings += check_live_temps()
+    findings.sort(key=lambda f: _ORDER.get(f.severity, 9))
+    return findings

src/rigdoctor/core/installer.py

@@ -0,0 +1,58 @@
+"""Optional-dependency installer (M9): figure out what's missing and install it.
+
+apt-only (D15). Installs run via pkexec/sudo so a normal user gets a single auth
+prompt; nothing is installed without an explicit confirmation by the caller.
+"""
+
+from __future__ import annotations
+
+import os
+import shlex
+import shutil
+import subprocess
+from collections.abc import Callable
+
+from . import sysenv
+from .catalog import COMPONENTS, Component
+
+
+def component_status(present: Callable[[str], bool] | None = None) -> list[tuple[Component, bool]]:
+    """Pair each catalog component with whether it's installed (command present)."""
+    present = present or sysenv.has_command
+    return [(c, present(c.command)) for c in COMPONENTS]
+
+
+def missing_packages(components: list[Component]) -> list[str]:
+    """De-duplicated apt package list for the given components, order preserved."""
+    packages: list[str] = []
+    for component in components:
+        for pkg in component.apt:
+            if pkg not in packages:
+                packages.append(pkg)
+    return packages
+
+
+def apt_install_command(packages: list[str]) -> list[str]:
+    """Build an `apt-get update && install` command, elevated if we're not root."""
+    inner = "apt-get update && apt-get install -y " + " ".join(shlex.quote(p) for p in packages)
+    cmd = ["/bin/sh", "-c", inner]
+    if os.geteuid() == 0:
+        return cmd
+    if shutil.which("pkexec"):
+        return ["pkexec", *cmd]
+    if shutil.which("sudo"):
+        return ["sudo", *cmd]
+    return cmd  # no privilege escalation available — will likely fail, surfaced to the caller
+
+
+def install_packages(packages: list[str]) -> tuple[int, str]:
+    """Install the given packages. Returns (exit_code, combined_output)."""
+    if not packages:
+        return (0, "Nothing to install.")
+    try:
+        proc = subprocess.run(
+            apt_install_command(packages), capture_output=True, text=True, timeout=900
+        )
+        return (proc.returncode, proc.stdout + proc.stderr)
+    except (subprocess.SubprocessError, OSError) as exc:
+        return (1, str(exc))

src/rigdoctor/core/inventory.py

@@ -0,0 +1,206 @@
+"""System inventory (M5): collect hardware/OS details, exportable to Markdown/JSON.
+
+Stdlib + tools already used elsewhere (nvidia-smi, lspci, lsblk, dmidecode). Every
+probe degrades gracefully; board/BIOS/RAM-module details need dmidecode as root.
+"""
+
+from __future__ import annotations
+
+import json
+import os
+import platform
+import shutil
+import subprocess
+from dataclasses import dataclass
+from pathlib import Path
+
+from .. import __version__
+from . import sysenv
+
+
+@dataclass
+class Section:
+    title: str
+    items: list[tuple[str, str]]
+
+
+def _run(cmd: list[str], timeout: float = 12.0) -> str:
+    try:
+        proc = subprocess.run(cmd, capture_output=True, text=True, timeout=timeout)
+        if proc.returncode == 0:
+            return proc.stdout
+    except (subprocess.SubprocessError, OSError):
+        pass
+    return ""
+
+
+def _system() -> Section:
+    u = os.uname()
+    return Section("System", [
+        ("Distro", sysenv.distro_name()),
+        ("Kernel", u.release),
+        ("Architecture", u.machine),
+        ("Hostname", u.nodename),
+        ("Python", platform.python_version()),
+        ("RigDoctor", __version__),
+    ])
+
+
+def _cpu() -> Section:
+    model = "?"
+    threads = 0
+    core_ids: set[tuple[str, str]] = set()
+    phys = "0"
+    try:
+        for line in Path("/proc/cpuinfo").read_text().splitlines():
+            if line.startswith("model name") and model == "?":
+                model = line.split(":", 1)[1].strip()
+            elif line.startswith("processor"):
+                threads += 1
+            elif line.startswith("physical id"):
+                phys = line.split(":", 1)[1].strip()
+            elif line.startswith("core id"):
+                core_ids.add((phys, line.split(":", 1)[1].strip()))
+    except OSError:
+        pass
+    items = [("Model", model)]
+    if core_ids:
+        items.append(("Cores", str(len(core_ids))))
+    items.append(("Threads", str(threads or os.cpu_count() or "?")))
+    return Section("CPU", items)
+
+
+def _firmware(dmi: dict) -> Section:
+    board = dmi.get("baseboard", {})
+    bios = dmi.get("bios", {})
+    items: list[tuple[str, str]] = []
+    if board:
+        items.append(("Motherboard", f"{board.get('Manufacturer', '')} {board.get('Product Name', '')}".strip()))
+    if bios:
+        items.append(("BIOS", f"{bios.get('Vendor', '')} {bios.get('Version', '')}".strip()))
+        if bios.get("Release Date"):
+            items.append(("BIOS date", bios["Release Date"]))
+    if not items:
+        items = [("Motherboard / BIOS", "run with admin (dmidecode needs root)")]
+    return Section("Firmware", items)
+
+
+def _memory(dmi: dict) -> Section:
+    items: list[tuple[str, str]] = []
+    try:
+        for line in Path("/proc/meminfo").read_text().splitlines():
+            if line.startswith("MemTotal"):
+                items.append(("Total", f"{int(line.split()[1]) / 1024 / 1024:.1f} GB"))
+                break
+    except (OSError, ValueError, IndexError):
+        pass
+    modules = dmi.get("memory", [])
+    if modules:
+        items.append(("Modules", str(len(modules))))
+        for i, m in enumerate(modules):
+            desc = " · ".join(p for p in (m.get("Size"), m.get("Type"), m.get("Speed"), m.get("Part Number")) if p)
+            items.append((f"Slot {i}", desc))
+    elif shutil.which("dmidecode"):
+        items.append(("Modules", "run with admin for module details"))
+    return Section("Memory", items)
+
+
+def _gpu() -> Section:
+    if shutil.which("nvidia-smi"):
+        out = _run([
+            "nvidia-smi",
+            "--query-gpu=name,driver_version,vbios_version,memory.total,pcie.link.gen.max,pcie.link.width.max",
+            "--format=csv,noheader",
+        ])
+        line = out.strip().splitlines()[0] if out.strip() else ""
+        if line:
+            cols = [c.strip() for c in line.split(",")]
+            keys = ["Name", "Driver", "VBIOS", "VRAM", "PCIe gen (max)", "PCIe width (max)"]
+            return Section("GPU", list(zip(keys, cols)))
+    out = _run(["lspci"])
+    gpus = [ln.split(":", 2)[-1].strip() for ln in out.splitlines()
+            if "VGA compatible controller" in ln or "3D controller" in ln]
+    return Section("GPU", [("Device", g) for g in gpus] or [("Device", "unknown")])
+
+
+def _storage() -> Section:
+    items: list[tuple[str, str]] = []
+    # TYPE first so MODEL (which can contain spaces) is the trailing field.
+    out = _run(["lsblk", "-dn", "-o", "TYPE,NAME,SIZE,MODEL"])
+    for line in out.strip().splitlines():
+        parts = line.split(None, 3)
+        if len(parts) < 3 or parts[0] != "disk":  # skip loop/zram/rom devices
+            continue
+        name, size = parts[1], parts[2]
+        model = parts[3] if len(parts) > 3 else ""
+        items.append((name, f"{model} ({size})".strip()))
+    return Section("Storage", items or [("Disks", "unknown")])
+
+
+def _display() -> Section:
+    return Section("Display", [
+        ("Session", os.environ.get("XDG_SESSION_TYPE", "unknown")),
+        ("Desktop", os.environ.get("XDG_CURRENT_DESKTOP") or os.environ.get("DESKTOP_SESSION", "unknown")),
+    ])
+
+
+def _dmidecode() -> dict:
+    if not shutil.which("dmidecode"):
+        return {}
+    out = _run(["dmidecode", "-t", "baseboard", "-t", "bios", "-t", "memory"], timeout=15)
+    if not out.strip():
+        return {}
+    result: dict = {"baseboard": {}, "bios": {}, "memory": []}
+    for block in out.split("Handle "):
+        lines = block.splitlines()
+        if len(lines) < 2:
+            continue
+        title = lines[1].strip()
+        kv: dict[str, str] = {}
+        for ln in lines[2:]:
+            if ln.startswith("\t") and ":" in ln:
+                key, _, value = ln.strip().partition(":")
+                kv[key.strip()] = value.strip()
+        if title == "Base Board Information":
+            result["baseboard"] = kv
+        elif title == "BIOS Information":
+            result["bios"] = kv
+        elif title == "Memory Device" and kv.get("Size") and kv["Size"] != "No Module Installed":
+            result["memory"].append(kv)
+    return result
+
+
+def collect() -> list[Section]:
+    from . import elevation
+
+    priv = elevation.privileged()
+    dmi = priv["dmidecode"] if (priv and priv.get("dmidecode") is not None) else _dmidecode()
+    return [_system(), _cpu(), _firmware(dmi), _memory(dmi), _gpu(), _storage(), _display()]
+
+
+def to_dict(sections: list[Section]) -> dict:
+    return {s.title: dict(s.items) for s in sections}
+
+
+def from_dict(data: dict) -> list[Section]:
+    return [Section(title, list(items.items())) for title, items in data.items()]
+
+
+def render_markdown(sections: list[Section]) -> str:
+    out = ["# RigDoctor system inventory", ""]
+    for s in sections:
+        out.append(f"## {s.title}")
+        out += [f"- **{k}:** {v}" for k, v in s.items]
+        out.append("")
+    return "\n".join(out).strip() + "\n"
+
+
+def render_text(sections: list[Section]) -> str:
+    blocks = []
+    for s in sections:
+        blocks.append("\n".join([s.title] + [f"  {k:<18} {v}" for k, v in s.items]))
+    return "\n\n".join(blocks)
+
+
+def render_json(sections: list[Section]) -> str:
+    return json.dumps(to_dict(sections), indent=2, ensure_ascii=False)

src/rigdoctor/core/sysenv.py

@@ -0,0 +1,49 @@
+"""Environment detection for the installer (M9)."""
+
+from __future__ import annotations
+
+import shutil
+import subprocess
+
+
+def package_manager() -> str | None:
+    """Only apt is supported (D15); return 'apt' if present, else None."""
+    if shutil.which("apt-get") or shutil.which("apt"):
+        return "apt"
+    return None
+
+
+def has_command(cmd: str) -> bool:
+    return shutil.which(cmd) is not None
+
+
+def distro_name() -> str:
+    try:
+        data: dict[str, str] = {}
+        with open("/etc/os-release") as f:
+            for line in f:
+                key, _, value = line.partition("=")
+                data[key.strip()] = value.strip().strip('"')
+        return data.get("PRETTY_NAME") or data.get("NAME") or "Linux"
+    except OSError:
+        return "Linux"
+
+
+def gpu_vendors() -> list[str]:
+    vendors: list[str] = []
+    if shutil.which("nvidia-smi"):
+        vendors.append("NVIDIA")
+    out = ""
+    if shutil.which("lspci"):
+        try:
+            out = subprocess.run(["lspci"], capture_output=True, text=True, timeout=10).stdout
+        except (subprocess.SubprocessError, OSError):
+            out = ""
+    low = out.lower()
+    if "nvidia" in low and "NVIDIA" not in vendors:
+        vendors.append("NVIDIA")
+    if ("amd/ati" in low or "advanced micro devices" in low or "radeon" in low) and "AMD" not in vendors:
+        vendors.append("AMD")
+    if "intel" in low and any(k in low for k in ("vga", "display", "graphics")) and "Intel" not in vendors:
+        vendors.append("Intel")
+    return vendors

src/rigdoctor/core/uninstall.py

@@ -0,0 +1,42 @@
+"""Uninstall the user-local RigDoctor install (app files; optionally all data).
+
+Mirrors `install.sh --uninstall`. The removal runs in a detached shell so it can
+delete the venv the current process is running from once we exit.
+"""
+
+from __future__ import annotations
+
+import shlex
+import subprocess
+from pathlib import Path
+
+from .. import config
+from . import reccontrol
+
+
+def targets(purge: bool = False) -> list[Path]:
+    """Paths removed by an uninstall. With purge, also config/state/logs."""
+    home = Path.home()
+    share = config.DATA_DIR.parent  # ~/.local/share
+    items = [
+        config.DATA_DIR / "venv",
+        home / ".local" / "bin" / "rigdoctor",
+        home / ".local" / "bin" / "rigdoctor-gui",
+        share / "applications" / "rigdoctor.desktop",
+        share / "icons" / "hicolor" / "scalable" / "apps" / "rigdoctor.svg",
+    ]
+    if purge:
+        items += [config.CONFIG_DIR, config.STATE_DIR, config.DATA_DIR]
+    return items
+
+
+def uninstall(purge: bool = False) -> None:
+    """Stop the recorder, clear the token if purging, and remove the install."""
+    reccontrol.stop_background()
+    if purge:
+        config.clear_token()  # removes keyring entry + any file fallback
+    paths = " ".join(shlex.quote(str(p)) for p in targets(purge))
+    subprocess.Popen(
+        ["/bin/sh", "-c", f"sleep 1; rm -rf {paths}"],
+        start_new_session=True,
+    )

src/rigdoctor/core/updates.py

@@ -0,0 +1,119 @@
+"""Update check (M13): ask the Gitea releases API for the latest version + notes.
+
+Stdlib-only (urllib). The Gitea instance requires sign-in, so updates are gated to
+account holders via a Personal Access Token (D18): set $RIGDOCTOR_TOKEN or save one
+with `rigdoctor login`. Returns the latest tag, its release notes (body), and a clear
+state for the UI; `apply_update` performs the no-root self-update.
+"""
+
+from __future__ import annotations
+
+import json
+import subprocess
+import sys
+import urllib.error
+import urllib.request
+
+from .. import __version__
+from ..config import load_token
+
+GITEA_BASE = "https://git.jesseyvanofferen.com"
+REPO = "jessey/rigdoctor"
+LATEST_API = f"{GITEA_BASE}/api/v1/repos/{REPO}/releases/latest"
+RELEASES_PAGE = f"{GITEA_BASE}/{REPO}/releases"
+TOKEN_PAGE = f"{GITEA_BASE}/user/settings/applications"
+
+# Update states
+NO_TOKEN = "no-token"
+AUTH = "auth"
+NETWORK = "network"
+UP_TO_DATE = "up-to-date"
+AVAILABLE = "available"
+
+
+def _parse(version: str) -> tuple[int, ...]:
+    return tuple(int(p) for p in version.lstrip("vV").split(".") if p.isdigit())
+
+
+def is_newer(latest: str, current: str = __version__) -> bool:
+    try:
+        return _parse(latest) > _parse(current)
+    except (ValueError, AttributeError):
+        return False
+
+
+def fetch_latest(timeout: float = 5.0) -> tuple[str | None, str, str | None]:
+    """Return (tag, notes, error). error is NO_TOKEN/AUTH/NETWORK, or None on success."""
+    token = load_token()
+    if not token:
+        return (None, "", NO_TOKEN)
+    req = urllib.request.Request(
+        LATEST_API,
+        headers={"Accept": "application/json", "Authorization": f"token {token}"},
+    )
+    try:
+        with urllib.request.urlopen(req, timeout=timeout) as resp:  # noqa: S310 (https)
+            data = json.load(resp)
+        return (data.get("tag_name") or None, (data.get("body") or "").strip(), None)
+    except urllib.error.HTTPError as exc:
+        return (None, "", AUTH if exc.code in (401, 403) else NETWORK)
+    except Exception:
+        return (None, "", NETWORK)
+
+
+def check_latest(timeout: float = 5.0) -> str | None:
+    """Convenience: latest tag or None (ignores notes/error)."""
+    tag, _notes, _error = fetch_latest(timeout)
+    return tag
+
+
+def update_state(timeout: float = 5.0) -> tuple[str, str | None, str]:
+    """Return (state, tag, notes). state in NO_TOKEN/AUTH/NETWORK/UP_TO_DATE/AVAILABLE."""
+    tag, notes, error = fetch_latest(timeout)
+    if error:
+        return (error, None, "")
+    if tag and is_newer(tag):
+        return (AVAILABLE, tag, notes)
+    return (UP_TO_DATE, tag, notes)
+
+
+def list_releases(limit: int = 15, timeout: float = 6.0) -> tuple[list[tuple[str, str, str]], str | None]:
+    """Return ([(tag, date, notes), …], error) for the in-app changelog."""
+    token = load_token()
+    if not token:
+        return ([], NO_TOKEN)
+    req = urllib.request.Request(
+        f"{GITEA_BASE}/api/v1/repos/{REPO}/releases?limit={limit}",
+        headers={"Accept": "application/json", "Authorization": f"token {token}"},
+    )
+    try:
+        with urllib.request.urlopen(req, timeout=timeout) as resp:  # noqa: S310 (https)
+            data = json.load(resp)
+        return ([
+            (r.get("tag_name") or "?", (r.get("published_at") or "")[:10], (r.get("body") or "").strip())
+            for r in data
+        ], None)
+    except urllib.error.HTTPError as exc:
+        return ([], AUTH if exc.code in (401, 403) else NETWORK)
+    except Exception:
+        return ([], NETWORK)
+
+
+def apply_update(tag: str) -> tuple[int, str]:
+    """Self-update the current (user-local) install to `tag` via authenticated pip.
+
+    Installs `rigdoctor[gui] @ git+https://oauth2:<token>@…/rigdoctor.git@<tag>` into
+    the running environment. Returns (exit_code, output) with the token scrubbed.
+    """
+    token = load_token()
+    if not token:
+        return (1, "No update token configured. Run `rigdoctor login`.")
+    host = GITEA_BASE.split("://", 1)[1]
+    ref = f"rigdoctor[gui] @ git+https://oauth2:{token}@{host}/{REPO}.git@{tag}"
+    cmd = [sys.executable, "-m", "pip", "install", "--upgrade", ref]
+    try:
+        proc = subprocess.run(cmd, capture_output=True, text=True, timeout=1800)
+        out = (proc.stdout + proc.stderr).replace(token, "***")
+        return (proc.returncode, out)
+    except (subprocess.SubprocessError, OSError) as exc:
+        return (1, str(exc).replace(token, "***"))

src/rigdoctor/gui/app.py

@@ -3,18 +3,28 @@
 from __future__ import annotations
 
 import sys
+from pathlib import Path
 
+from PySide6.QtGui import QIcon
 from PySide6.QtWidgets import QApplication
 
 from ..config import load_config
+from . import desktop
 from .main_window import MainWindow
 from .theme import STYLESHEET
 
+ICON = Path(__file__).parent / "assets" / "rigdoctor.svg"
+
 
 def main(argv: list[str] | None = None) -> int:
+    desktop.ensure()  # self-register icon + .desktop so updates show it without re-installing
     app = QApplication(argv if argv is not None else sys.argv)
     app.setApplicationName("RigDoctor")
     app.setApplicationDisplayName("RigDoctor")
+    # Match the installed rigdoctor.desktop so the dock/launcher shows our icon (Wayland app-id).
+    app.setDesktopFileName("rigdoctor")
+    if ICON.exists():
+        app.setWindowIcon(QIcon(str(ICON)))
     app.setStyle("Fusion")
     app.setStyleSheet(STYLESHEET)
 

src/rigdoctor/gui/assets/rigdoctor.svg

@@ -0,0 +1,8 @@
+<svg xmlns="http://www.w3.org/2000/svg" width="256" height="256" viewBox="0 0 256 256">
+  <rect x="8" y="8" width="240" height="240" rx="52" fill="#15181e"/>
+  <circle cx="128" cy="128" r="84" fill="none" stroke="#2a2f39" stroke-width="14"/>
+  <path d="M128 44 a84 84 0 1 1 -59.4 24.6" fill="none" stroke="#38bdf8"
+        stroke-width="14" stroke-linecap="round"/>
+  <path d="M60 132 H100 L116 96 L140 168 L156 132 H196" fill="none" stroke="#e6e8eb"
+        stroke-width="14" stroke-linecap="round" stroke-linejoin="round"/>
+</svg>

src/rigdoctor/gui/desktop.py

@@ -0,0 +1,51 @@
+"""Best-effort desktop integration: install our icon + .desktop so the dock shows it.
+
+Runs at GUI launch (idempotent), so a self-update + relaunch refreshes the icon without
+re-running install.sh. No-op for non-installed (dev) runs where the launcher is absent.
+"""
+
+from __future__ import annotations
+
+import shutil
+import sys
+from pathlib import Path
+
+from .. import config
+
+_ICON_SRC = Path(__file__).parent / "assets" / "rigdoctor.svg"
+
+_DESKTOP = """[Desktop Entry]
+Type=Application
+Name=RigDoctor
+Comment=Hardware monitoring & crash diagnostics for Linux gamers
+Exec={exec}
+Icon=rigdoctor
+Terminal=false
+Categories=System;Monitor;Utility;
+StartupWMClass=rigdoctor
+"""
+
+
+def ensure() -> None:
+    share = config.DATA_DIR.parent  # ~/.local/share
+
+    try:
+        if _ICON_SRC.exists():
+            icon_dst = share / "icons" / "hicolor" / "scalable" / "apps" / "rigdoctor.svg"
+            icon_dst.parent.mkdir(parents=True, exist_ok=True)
+            if not icon_dst.exists() or icon_dst.read_bytes() != _ICON_SRC.read_bytes():
+                shutil.copyfile(_ICON_SRC, icon_dst)
+    except OSError:
+        pass
+
+    gui_exec = Path(sys.executable).with_name("rigdoctor-gui")
+    if not gui_exec.exists():  # dev / not a normal install — don't fabricate a .desktop
+        return
+    try:
+        desktop = share / "applications" / "rigdoctor.desktop"
+        content = _DESKTOP.format(exec=gui_exec)
+        desktop.parent.mkdir(parents=True, exist_ok=True)
+        if not desktop.exists() or desktop.read_text() != content:
+            desktop.write_text(content)
+    except OSError:
+        pass

src/rigdoctor/gui/health_page.py

@@ -0,0 +1,129 @@
+"""Health page (M4 in the GUI): runs the health checks and shows findings as cards."""
+
+from __future__ import annotations
+
+import threading
+import time
+
+from PySide6.QtCore import Qt, QTimer, Signal
+from PySide6.QtWidgets import (
+    QFrame,
+    QHBoxLayout,
+    QLabel,
+    QPushButton,
+    QScrollArea,
+    QVBoxLayout,
+    QWidget,
+)
+
+from .theme import ACCENT, CRIT, GOOD, MUTED, WARN
+
+_SEV = {
+    "critical": ("CRITICAL", CRIT),
+    "warning": ("WARNING", WARN),
+    "info": ("INFO", MUTED),
+    "ok": ("OK", GOOD),
+}
+
+
+def _finding_widget(finding) -> QFrame:
+    label, color = _SEV.get(finding.severity, ("?", MUTED))
+    card = QFrame()
+    card.setObjectName("Card")
+    v = QVBoxLayout(card)
+    v.setContentsMargins(16, 12, 16, 12)
+    v.setSpacing(4)
+
+    head = QLabel(f"{label}  ·  {finding.category}: {finding.title}")
+    head.setStyleSheet(f"color: {color}; font-weight: 700; background: transparent;")
+    head.setWordWrap(True)
+    v.addWidget(head)
+
+    if finding.detail:
+        detail = QLabel(finding.detail)
+        detail.setObjectName("Muted")
+        detail.setWordWrap(True)
+        v.addWidget(detail)
+    if finding.suggestion:
+        suggestion = QLabel(f"→ {finding.suggestion}")
+        suggestion.setStyleSheet(f"color: {ACCENT}; background: transparent;")
+        suggestion.setWordWrap(True)
+        v.addWidget(suggestion)
+    return card
+
+
+class HealthPage(QWidget):
+    _result = Signal(object)  # list[Finding]
+
+    def __init__(self) -> None:
+        super().__init__()
+        self.setObjectName("Page")
+        self._result.connect(self._render_findings)
+
+        root = QVBoxLayout(self)
+        root.setContentsMargins(20, 18, 20, 18)
+        root.setSpacing(16)
+
+        header = QHBoxLayout()
+        title = QLabel("Health")
+        title.setObjectName("PageTitle")
+        header.addWidget(title)
+        header.addStretch(1)
+        self._status = QLabel("")
+        self._status.setObjectName("Muted")
+        header.addWidget(self._status)
+        self._run_btn = QPushButton("Run health report")
+        self._run_btn.setObjectName("PrimaryButton")
+        self._run_btn.clicked.connect(self._run)
+        header.addWidget(self._run_btn)
+        root.addLayout(header)
+
+        scroll = QScrollArea()
+        scroll.setWidgetResizable(True)
+        scroll.setFrameShape(QFrame.Shape.NoFrame)
+        scroll.setStyleSheet("background: transparent;")
+        self._container = QWidget()
+        self._list = QVBoxLayout(self._container)
+        self._list.setContentsMargins(0, 0, 0, 0)
+        self._list.setSpacing(10)
+        self._list.setAlignment(Qt.AlignmentFlag.AlignTop)
+        scroll.setWidget(self._container)
+        root.addWidget(scroll, 1)
+
+        QTimer.singleShot(300, self._run)  # auto-run shortly after the window opens
+
+    def _run(self) -> None:
+        self._run_btn.setEnabled(False)
+        self._status.setText("Scanning logs, SMART, and driver…")
+        threading.Thread(target=self._work, daemon=True).start()
+
+    def _work(self) -> None:
+        from ..core.health import run_health_checks
+
+        try:
+            findings = run_health_checks()
+        except Exception:
+            findings = []
+        self._result.emit(findings)
+
+    def _render_findings(self, findings) -> None:
+        self._run_btn.setEnabled(True)
+        if findings is None:  # collection failed — keep current results
+            self._status.setText("check failed")
+            return
+
+        while self._list.count():
+            item = self._list.takeAt(0)
+            w = item.widget()
+            if w is not None:
+                w.deleteLater()
+
+        crit = sum(1 for f in findings if f.severity == "critical")
+        warn = sum(1 for f in findings if f.severity == "warning")
+        self._status.setText(
+            f"{crit} critical · {warn} warning · {len(findings)} checks  ·  "
+            f"{time.strftime('%H:%M:%S')}"
+        )
+        for finding in findings:
+            self._list.addWidget(_finding_widget(finding))
+        self._list.addStretch(1)

src/rigdoctor/gui/inventory_page.py

@@ -0,0 +1,145 @@
+"""Inventory page (M5 in the GUI): system inventory with copy/save + admin re-collect."""
+
+from __future__ import annotations
+
+import os
+import threading
+
+from PySide6.QtCore import Qt, QTimer, Signal
+from PySide6.QtWidgets import (
+    QApplication,
+    QFileDialog,
+    QFrame,
+    QGridLayout,
+    QHBoxLayout,
+    QLabel,
+    QPushButton,
+    QScrollArea,
+    QVBoxLayout,
+    QWidget,
+)
+
+from ..core import inventory
+from .theme import MUTED
+
+
+def _section_card(section) -> QFrame:
+    card = QFrame()
+    card.setObjectName("Card")
+    layout = QVBoxLayout(card)
+    layout.setContentsMargins(16, 12, 16, 12)
+    layout.setSpacing(6)
+    title = QLabel(section.title)
+    title.setStyleSheet("font-weight: 700; background: transparent;")
+    layout.addWidget(title)
+    grid = QGridLayout()
+    grid.setColumnStretch(1, 1)
+    grid.setHorizontalSpacing(14)
+    grid.setVerticalSpacing(4)
+    for row, (key, value) in enumerate(section.items):
+        k = QLabel(key)
+        k.setObjectName("Muted")
+        v = QLabel(value)
+        v.setWordWrap(True)
+        v.setStyleSheet("background: transparent;")
+        grid.addWidget(k, row, 0)
+        grid.addWidget(v, row, 1)
+    layout.addLayout(grid)
+    return card
+
+
+class InventoryPage(QWidget):
+    _result = Signal(object)  # list[Section]
+
+    def __init__(self) -> None:
+        super().__init__()
+        self.setObjectName("Page")
+        self._sections: list = []
+        self._result.connect(self._render)
+
+        root = QVBoxLayout(self)
+        root.setContentsMargins(20, 18, 20, 18)
+        root.setSpacing(16)
+
+        header = QHBoxLayout()
+        title = QLabel("Inventory")
+        title.setObjectName("PageTitle")
+        header.addWidget(title)
+        header.addStretch(1)
+        self._status = QLabel("")
+        self._status.setObjectName("Muted")
+        header.addWidget(self._status)
+        self._copy_btn = QPushButton("Copy Markdown")
+        self._copy_btn.clicked.connect(self._copy)
+        header.addWidget(self._copy_btn)
+        self._save_btn = QPushButton("Save…")
+        self._save_btn.clicked.connect(self._save)
+        header.addWidget(self._save_btn)
+        self._refresh_btn = QPushButton("Refresh")
+        self._refresh_btn.setObjectName("PrimaryButton")
+        self._refresh_btn.clicked.connect(self._run)
+        header.addWidget(self._refresh_btn)
+        root.addLayout(header)
+
+        scroll = QScrollArea()
+        scroll.setWidgetResizable(True)
+        scroll.setFrameShape(QFrame.Shape.NoFrame)
+        scroll.setStyleSheet("background: transparent;")
+        self._container = QWidget()
+        self._list = QVBoxLayout(self._container)
+        self._list.setContentsMargins(0, 0, 0, 0)
+        self._list.setSpacing(12)
+        self._list.setAlignment(Qt.AlignmentFlag.AlignTop)
+        scroll.setWidget(self._container)
+        root.addWidget(scroll, 1)
+
+        QTimer.singleShot(300, self._run)
+
+    def _run(self) -> None:
+        self._busy("Collecting…")
+        threading.Thread(target=self._work, daemon=True).start()
+
+    def _work(self) -> None:
+        try:
+            sections = inventory.collect()
+        except Exception:
+            sections = []
+        self._result.emit(sections)
+
+    def _busy(self, text: str) -> None:
+        self._status.setText(text)
+        for b in (self._refresh_btn, self._copy_btn, self._save_btn):
+            b.setEnabled(False)
+
+    def _render(self, sections) -> None:
+        self._refresh_btn.setEnabled(True)
+        self._copy_btn.setEnabled(True)
+        self._save_btn.setEnabled(True)
+        if sections is None:  # collection failed — keep current
+            self._status.setText("collection failed")
+            return
+
+        self._sections = sections
+        while self._list.count():
+            item = self._list.takeAt(0)
+            w = item.widget()
+            if w is not None:
+                w.deleteLater()
+        for section in sections:
+            self._list.addWidget(_section_card(section))
+        self._list.addStretch(1)
+        self._status.setText("")
+
+    def _copy(self) -> None:
+        if self._sections:
+            QApplication.clipboard().setText(inventory.render_markdown(self._sections))
+            self._status.setText("copied as Markdown")
+
+    def _save(self) -> None:
+        if not self._sections:
+            return
+        path, _ = QFileDialog.getSaveFileName(self, "Save inventory", "rigdoctor-inventory.md", "Markdown (*.md)")
+        if path:
+            with open(path, "w", encoding="utf-8") as f:
+                f.write(inventory.render_markdown(self._sections))
+            self._status.setText(f"saved {os.path.basename(path)}")

src/rigdoctor/gui/main_window.py

@@ -2,36 +2,54 @@
 
 from __future__ import annotations
 
-from PySide6.QtCore import Qt
+import os
+import sys
+import threading
+
+from PySide6.QtCore import Qt, QProcess, QTimer, Signal
+from PySide6.QtGui import QTextDocument
 from PySide6.QtWidgets import (
+    QApplication,
     QButtonGroup,
+    QDialog,
     QFrame,
     QHBoxLayout,
     QLabel,
     QMainWindow,
+    QMessageBox,
     QPushButton,
     QStackedWidget,
+    QTextEdit,
     QVBoxLayout,
     QWidget,
 )
 
+from .. import __version__
+from ..config import load_config
+from ..core import alerts, elevation, updates
 from .dashboard import Dashboard
+from .health_page import HealthPage
+from .inventory_page import InventoryPage
+from .notifications_page import NotificationsPage
 from .recorder_page import RecorderPage
-from .theme import ACCENT, MUTED
+from .setup_page import SetupPage
+from .theme import ACCENT, GOOD, MUTED
 from .worker import SamplerWorker
 
-_NAV_ITEMS = ["Dashboard", "Logs", "Health", "Inventory"]
-_PLACEHOLDERS = {
-    "Health": "The health report (M4) — log scan + plain-language findings — lands here.",
-    "Inventory": "System inventory (M5) — CPU/GPU/board/RAM/drivers — lands here.",
-}
+_NAV_ITEMS = ["Dashboard", "Logs", "Health", "Setup", "Inventory", "Notifications"]
 
 
 class MainWindow(QMainWindow):
+    _update_checked = Signal(object)  # (state, tag, notes)
+    _update_applied = Signal(int)     # pip exit code
+    _changelog_ready = Signal(object)  # ([(tag, date, notes)], error)
+    _elevated = Signal()              # privileged data collected at launch
+
     def __init__(self, interval: float = 1.0) -> None:
         super().__init__()
         self.setWindowTitle("RigDoctor")
         self.resize(1000, 680)
+        cfg = load_config()
 
         central = QWidget()
         self.setCentralWidget(central)
@@ -47,10 +65,17 @@ class MainWindow(QMainWindow):
         self._stack = QStackedWidget()
         self.dashboard = Dashboard()
         self.recorder_page = RecorderPage()
+        self.health_page = HealthPage()
+        self.setup_page = SetupPage()
+        self.inventory_page = InventoryPage()
+        self.notifications_page = NotificationsPage()
+        self.notifications_page.changed.connect(self._apply_alert_settings)
         self._stack.addWidget(self.dashboard)            # 0 Dashboard
         self._stack.addWidget(self.recorder_page)        # 1 Logs
-        self._stack.addWidget(self._placeholder_page("Health", _PLACEHOLDERS["Health"]))      # 2
-        self._stack.addWidget(self._placeholder_page("Inventory", _PLACEHOLDERS["Inventory"]))  # 3
+        self._stack.addWidget(self.health_page)          # 2 Health
+        self._stack.addWidget(self.setup_page)           # 3 Setup
+        self._stack.addWidget(self.inventory_page)       # 4 Inventory
+        self._stack.addWidget(self.notifications_page)   # 5 Notifications
         content_layout.addWidget(self._stack)
 
         layout.addWidget(self._build_sidebar())
@@ -58,8 +83,40 @@ class MainWindow(QMainWindow):
 
         self._worker = SamplerWorker(interval=interval)
         self._worker.sampled.connect(self.dashboard.update_sample)
+        # Desktop alerts (M8): overheat / GPU-lost from the sample stream, new-version below.
+        # Configurable on the Notifications page; gated by AlertMonitor.enabled.
+        self._notified_update_tag = None
+        self._alert_monitor = alerts.AlertMonitor(
+            gpu_temp=float(cfg.get("gpu_temp_alert", 90.0)),
+            cpu_temp=float(cfg.get("cpu_temp_alert", 95.0)),
+        )
+        self._alert_monitor.enabled = bool(cfg.get("alerts_enabled", True))
+        self._worker.sampled.connect(self._alert_monitor.check)
         self._worker.start()
 
+        # Ask for the password once at launch and collect root-only data (SMART +
+        # dmidecode); Health/Inventory then always show the full picture (config:
+        # elevate_on_launch). Falls back silently to non-root if cancelled/unavailable.
+        if cfg.get("elevate_on_launch", True) and elevation.available():
+            self._elevated.connect(self._on_elevated)
+            threading.Thread(target=self._collect_privileged, daemon=True).start()
+
+        # Update check (M13): once at launch, then periodically so a newly published
+        # release is detected without restarting (interval from config; 0 disables).
+        self._latest_tag = None
+        self._latest_notes = ""
+        self._applied = False
+        self._update_checked.connect(self._show_update_state)
+        self._update_applied.connect(self._on_update_applied)
+        self._changelog_ready.connect(self._on_changelog)
+        self._start_update_check()
+        minutes = float(cfg.get("update_check_minutes", 30) or 0)
+        if minutes > 0:
+            self._update_timer = QTimer(self)
+            self._update_timer.setInterval(int(minutes * 60_000))
+            self._update_timer.timeout.connect(self._start_update_check)
+            self._update_timer.start()
+
     def _build_sidebar(self) -> QFrame:
         bar = QFrame()
         bar.setObjectName("Sidebar")
@@ -91,30 +148,161 @@ class MainWindow(QMainWindow):
         v.addStretch(1)
         live = QLabel(f'<span style="color:{ACCENT};">●</span> <span style="color:{MUTED};">Live</span>')
         v.addWidget(live)
+        version = QLabel(f"v{__version__}")
+        version.setObjectName("Muted")
+        v.addWidget(version)
+        changelog_btn = QPushButton("Changelog")
+        changelog_btn.setObjectName("LinkButton")
+        changelog_btn.setCursor(Qt.CursorShape.PointingHandCursor)
+        changelog_btn.clicked.connect(self._show_changelog)
+        v.addWidget(changelog_btn)
+        check_btn = QPushButton("Check for updates")
+        check_btn.setObjectName("LinkButton")
+        check_btn.setCursor(Qt.CursorShape.PointingHandCursor)
+        check_btn.clicked.connect(self._manual_check)
+        v.addWidget(check_btn)
+
+        # Update state (filled in by the background check).
+        self._update_label = QLabel("checking for updates…")
+        self._update_label.setObjectName("Muted")
+        v.addWidget(self._update_label)
+        self._update_btn = QPushButton()
+        self._update_btn.setObjectName("PrimaryButton")
+        self._update_btn.setCursor(Qt.CursorShape.PointingHandCursor)
+        self._update_btn.clicked.connect(self._apply_update)
+        self._update_btn.setVisible(False)
+        v.addWidget(self._update_btn)
+        self._restart_btn = QPushButton("Restart now")
+        self._restart_btn.setObjectName("PrimaryButton")
+        self._restart_btn.setCursor(Qt.CursorShape.PointingHandCursor)
+        self._restart_btn.clicked.connect(self._restart)
+        self._restart_btn.setVisible(False)
+        v.addWidget(self._restart_btn)
         return bar
 
-    def _placeholder_page(self, title: str, description: str) -> QWidget:
-        page = QWidget()
-        page.setObjectName("Page")
-        v = QVBoxLayout(page)
-        v.setContentsMargins(20, 18, 20, 18)
-        v.setSpacing(16)
-        head = QLabel(title)
-        head.setObjectName("PageTitle")
-        v.addWidget(head)
+    def _restart(self) -> None:
+        gui = os.path.join(os.path.dirname(sys.executable), "rigdoctor-gui")
+        if os.path.exists(gui):
+            QProcess.startDetached(gui)
+        else:  # dev / not installed next to python
+            QProcess.startDetached(sys.executable, sys.argv)
+        QApplication.instance().quit()
 
-        card = QFrame()
-        card.setObjectName("Card")
-        cv = QVBoxLayout(card)
-        cv.setContentsMargins(24, 48, 24, 48)
-        msg = QLabel(description)
-        msg.setObjectName("Muted")
-        msg.setWordWrap(True)
-        msg.setAlignment(Qt.AlignmentFlag.AlignCenter)
-        cv.addWidget(msg)
-        v.addWidget(card)
-        v.addStretch(1)
-        return page
+    def _apply_update(self) -> None:
+        if not self._latest_tag:
+            return
+        box = QMessageBox(self)
+        box.setWindowTitle(f"Update to {self._latest_tag}")
+        box.setText(f"Update RigDoctor to {self._latest_tag}?")
+        notes_doc = QTextDocument()
+        notes_doc.setMarkdown(self._latest_notes or "_(no release notes)_")
+        box.setInformativeText(notes_doc.toHtml())  # render Markdown as rich text (#1)
+        box.setStandardButtons(QMessageBox.StandardButton.Ok | QMessageBox.StandardButton.Cancel)
+        box.button(QMessageBox.StandardButton.Ok).setText("Update")
+        if box.exec() != QMessageBox.StandardButton.Ok:
+            return
+        self._update_btn.setEnabled(False)
+        self._update_label.setText("updating…")
+        tag = self._latest_tag
+        threading.Thread(target=lambda: self._update_applied.emit(updates.apply_update(tag)[0]), daemon=True).start()
+
+    def _on_update_applied(self, rc: int) -> None:
+        if rc == 0:
+            self._applied = True
+            self._update_label.setText("update installed")
+            self._update_btn.setVisible(False)
+            self._restart_btn.setVisible(True)
+            if hasattr(self, "_update_timer"):
+                self._update_timer.stop()
+        else:
+            self._update_label.setText("update failed")
+            self._update_btn.setEnabled(True)
+
+    def _collect_privileged(self) -> None:
+        data = elevation.collect_via_pkexec()
+        if data is not None:
+            elevation.set_privileged(data)
+            self._elevated.emit()
+
+    def _on_elevated(self) -> None:
+        # Re-run Health and Inventory now that root-only data (SMART/dmidecode) is available.
+        self.health_page._run()
+        self.inventory_page._run()
+
+    def _apply_alert_settings(self) -> None:
+        cfg = load_config()
+        self._alert_monitor.enabled = bool(cfg.get("alerts_enabled", True))
+        self._alert_monitor.gpu_temp = float(cfg.get("gpu_temp_alert", 90.0))
+        self._alert_monitor.cpu_temp = float(cfg.get("cpu_temp_alert", 95.0))
+
+    def _manual_check(self) -> None:
+        if self._applied:
+            return
+        self._update_label.setText("checking for updates…")
+        self._start_update_check()
+
+    def _start_update_check(self) -> None:
+        threading.Thread(target=self._check_updates, daemon=True).start()
+
+    def _show_changelog(self) -> None:
+        dialog = QDialog(self)
+        dialog.setWindowTitle("RigDoctor — Changelog")
+        dialog.resize(560, 540)
+        layout = QVBoxLayout(dialog)
+        view = QTextEdit()
+        view.setObjectName("Report")
+        view.setReadOnly(True)
+        view.setPlainText("Loading…")
+        layout.addWidget(view)
+        self._changelog_view = view
+        dialog.show()
+        threading.Thread(target=self._fetch_changelog, daemon=True).start()
+
+    def _fetch_changelog(self) -> None:
+        self._changelog_ready.emit(updates.list_releases())
+
+    def _on_changelog(self, result) -> None:
+        view = getattr(self, "_changelog_view", None)
+        if view is None:
+            return
+        releases, error = result
+        if error == updates.NO_TOKEN:
+            view.setPlainText("Add an update token (Setup → Update access) to load the changelog.")
+            return
+        if error or not releases:
+            view.setPlainText("Couldn't load the changelog from the update server.")
+            return
+        blocks = []
+        for tag, date, notes in releases:
+            title = f"## {tag}" + (f" — {date}" if date else "")
+            blocks.append(f"{title}\n\n{notes or '_(no notes)_'}")
+        view.setMarkdown("\n\n".join(blocks))  # render Markdown instead of raw text (#1)
+
+    def _check_updates(self) -> None:
+        self._update_checked.emit(updates.update_state())
+
+    def _show_update_state(self, result) -> None:
+        if self._applied:  # an update was applied this session; awaiting restart
+            return
+        state, tag, notes = result
+        self._latest_tag = tag
+        self._latest_notes = notes
+        self._update_btn.setVisible(False)
+        if state == updates.NO_TOKEN:
+            self._update_label.setText("connect to update server")
+        elif state == updates.AUTH:
+            self._update_label.setText("update access denied")
+        elif state == updates.NETWORK:
+            self._update_label.setText("update check unavailable")
+        elif state == updates.AVAILABLE:
+            self._update_label.setText(f'<span style="color:{GOOD};">{tag} available</span>')
+            self._update_btn.setText(f"Update to {tag}")
+            self._update_btn.setVisible(True)
+            if self._alert_monitor.enabled and tag != self._notified_update_tag:
+                self._notified_update_tag = tag  # once per version, not every poll
+                alerts.notify("Update available", f"RigDoctor {tag} is available — open RigDoctor to update.")
+        else:  # UP_TO_DATE
+            self._update_label.setText("up-to-date")
 
     def closeEvent(self, event) -> None:  # noqa: N802 (Qt override)
         self._worker.stop()

src/rigdoctor/gui/notifications_page.py

@@ -0,0 +1,108 @@
+"""Notifications page (M8 config): user-configurable alert settings."""
+
+from __future__ import annotations
+
+from PySide6.QtCore import Qt, Signal
+from PySide6.QtWidgets import (
+    QCheckBox,
+    QDoubleSpinBox,
+    QFrame,
+    QGridLayout,
+    QHBoxLayout,
+    QLabel,
+    QPushButton,
+    QVBoxLayout,
+    QWidget,
+)
+
+from ..config import load_config, update_config
+from ..core import alerts
+
+
+class NotificationsPage(QWidget):
+    changed = Signal()  # settings saved — main window re-applies them live
+
+    def __init__(self) -> None:
+        super().__init__()
+        self.setObjectName("Page")
+        root = QVBoxLayout(self)
+        root.setContentsMargins(20, 18, 20, 18)
+        root.setSpacing(16)
+
+        title = QLabel("Notifications")
+        title.setObjectName("PageTitle")
+        root.addWidget(title)
+
+        card = QFrame()
+        card.setObjectName("Card")
+        v = QVBoxLayout(card)
+        v.setContentsMargins(16, 14, 16, 14)
+        v.setSpacing(10)
+        head = QLabel("Alerts")
+        head.setStyleSheet("font-weight: 700; background: transparent;")
+        v.addWidget(head)
+
+        self._enabled = QCheckBox("Enable desktop notifications")
+        v.addWidget(self._enabled)
+
+        grid = QGridLayout()
+        grid.setHorizontalSpacing(12)
+        grid.setColumnStretch(2, 1)
+        self._gpu = self._spin()
+        self._cpu = self._spin()
+        grid.addWidget(QLabel("GPU temperature alert"), 0, 0)
+        grid.addWidget(self._gpu, 0, 1)
+        grid.addWidget(QLabel("CPU temperature alert"), 1, 0)
+        grid.addWidget(self._cpu, 1, 1)
+        v.addLayout(grid)
+
+        note = QLabel("GPU-lost and new-version alerts are included whenever notifications are enabled.")
+        note.setObjectName("Muted")
+        note.setWordWrap(True)
+        v.addWidget(note)
+
+        buttons = QHBoxLayout()
+        save = QPushButton("Save")
+        save.setObjectName("PrimaryButton")
+        save.clicked.connect(self._save)
+        test = QPushButton("Send test")
+        test.clicked.connect(self._test)
+        buttons.addWidget(save)
+        buttons.addWidget(test)
+        buttons.addStretch(1)
+        v.addLayout(buttons)
+        self._status = QLabel("")
+        self._status.setObjectName("Muted")
+        v.addWidget(self._status)
+
+        root.addWidget(card)
+        root.addStretch(1)
+        self._load()
+
+    @staticmethod
+    def _spin() -> QDoubleSpinBox:
+        spin = QDoubleSpinBox()
+        spin.setRange(40, 110)
+        spin.setDecimals(0)
+        spin.setSingleStep(1)
+        spin.setSuffix(" °C")
+        return spin
+
+    def _load(self) -> None:
+        cfg = load_config()
+        self._enabled.setChecked(bool(cfg.get("alerts_enabled", True)))
+        self._gpu.setValue(float(cfg.get("gpu_temp_alert", 90.0)))
+        self._cpu.setValue(float(cfg.get("cpu_temp_alert", 95.0)))
+
+    def _save(self) -> None:
+        update_config(
+            alerts_enabled=self._enabled.isChecked(),
+            gpu_temp_alert=self._gpu.value(),
+            cpu_temp_alert=self._cpu.value(),
+        )
+        self.changed.emit()
+        self._status.setText("Saved.")
+
+    def _test(self) -> None:
+        ok = alerts.notify("RigDoctor", "Test notification — alerts are working.")
+        self._status.setText("Test notification sent." if ok else "notify-send not found — install libnotify-bin (Setup).")

src/rigdoctor/gui/setup_page.py

@@ -0,0 +1,234 @@
+"""Setup page (M9 in the GUI): show environment + optional components, install missing."""
+
+from __future__ import annotations
+
+import threading
+
+from PySide6.QtCore import Qt, QUrl, Signal
+from PySide6.QtGui import QDesktopServices
+from PySide6.QtWidgets import (
+    QApplication,
+    QFrame,
+    QHBoxLayout,
+    QLabel,
+    QLineEdit,
+    QMessageBox,
+    QPushButton,
+    QSizePolicy,
+    QTextEdit,
+    QVBoxLayout,
+    QWidget,
+)
+
+from .. import config
+from ..core import installer, sysenv, uninstall, updates
+from .theme import GOOD, MUTED, WARN
+
+
+def _panel(title: str) -> tuple[QFrame, QVBoxLayout]:
+    frame = QFrame()
+    frame.setObjectName("Card")
+    frame.setSizePolicy(QSizePolicy.Policy.Expanding, QSizePolicy.Policy.Maximum)
+    layout = QVBoxLayout(frame)
+    layout.setContentsMargins(16, 14, 16, 14)
+    layout.setSpacing(8)
+    label = QLabel(title)
+    label.setStyleSheet("font-weight: 700; background: transparent;")
+    layout.addWidget(label)
+    return frame, layout
+
+
+_BACKEND_DESC = {
+    "env": "token from $RIGDOCTOR_TOKEN",
+    "keyring": "token stored in the OS keyring (encrypted)",
+    "file": "token stored in a 0600 file — install libsecret-tools to encrypt it",
+    "none": "no token saved",
+}
+
+
+class SetupPage(QWidget):
+    _installed = Signal(int, str)
+    _upd_state = Signal(object)
+
+    def __init__(self) -> None:
+        super().__init__()
+        self.setObjectName("Page")
+        self._installed.connect(self._on_installed)
+        self._upd_state.connect(self._on_upd_state)
+
+        root = QVBoxLayout(self)
+        root.setContentsMargins(20, 18, 20, 18)
+        root.setSpacing(16)
+
+        title = QLabel("Setup")
+        title.setObjectName("PageTitle")
+        root.addWidget(title)
+
+        env_card, env_layout = _panel("Environment")
+        self._env = QLabel("")
+        self._env.setObjectName("Muted")
+        env_layout.addWidget(self._env)
+        root.addWidget(env_card)
+
+        comp_card, comp_layout = _panel("Optional components")
+        self._components = QVBoxLayout()
+        self._components.setSpacing(6)
+        comp_layout.addLayout(self._components)
+        controls = QHBoxLayout()
+        self._install_btn = QPushButton("Install missing")
+        self._install_btn.setObjectName("PrimaryButton")
+        self._install_btn.clicked.connect(self._install)
+        self._refresh_btn = QPushButton("Re-check")
+        self._refresh_btn.clicked.connect(self._refresh)
+        controls.addWidget(self._install_btn)
+        controls.addWidget(self._refresh_btn)
+        controls.addStretch(1)
+        comp_layout.addLayout(controls)
+        root.addWidget(comp_card)
+
+        # Update access (M13): token gating updates to Gitea account holders.
+        upd_card, upd_layout = _panel("Update access")
+        self._upd_status = QLabel("")
+        self._upd_status.setObjectName("Muted")
+        self._upd_status.setWordWrap(True)
+        upd_layout.addWidget(self._upd_status)
+        token_row = QHBoxLayout()
+        self._token_input = QLineEdit()
+        self._token_input.setEchoMode(QLineEdit.EchoMode.Password)
+        self._token_input.setPlaceholderText("Paste a Gitea token (scope: read:repository)")
+        save_btn = QPushButton("Save token")
+        save_btn.setObjectName("PrimaryButton")
+        save_btn.clicked.connect(self._save_token)
+        get_btn = QPushButton("Get a token")
+        get_btn.clicked.connect(lambda: QDesktopServices.openUrl(QUrl(updates.TOKEN_PAGE)))
+        token_row.addWidget(self._token_input, 1)
+        token_row.addWidget(save_btn)
+        token_row.addWidget(get_btn)
+        upd_layout.addLayout(token_row)
+        root.addWidget(upd_card)
+
+        self._output = QTextEdit()
+        self._output.setObjectName("Report")
+        self._output.setReadOnly(True)
+        self._output.setMinimumHeight(180)
+        self._output.setVisible(False)
+        root.addWidget(self._output)
+        root.addStretch(1)
+
+        danger = QHBoxLayout()
+        danger.addStretch(1)
+        uninstall_btn = QPushButton("Uninstall RigDoctor")
+        uninstall_btn.setObjectName("DangerButton")
+        uninstall_btn.clicked.connect(self._uninstall)
+        danger.addWidget(uninstall_btn)
+        root.addLayout(danger)
+
+        self._refresh()
+        self._refresh_update_status()
+
+    def _uninstall(self) -> None:
+        box = QMessageBox(self)
+        box.setIcon(QMessageBox.Icon.Warning)
+        box.setWindowTitle("Uninstall RigDoctor")
+        box.setText("Uninstall RigDoctor?")
+        box.setInformativeText(
+            "This removes the app. Choose “Remove all” to also delete your settings, "
+            "update token, and captured logs."
+        )
+        remove_all = box.addButton("Remove all", QMessageBox.ButtonRole.DestructiveRole)
+        app_only = box.addButton("Uninstall", QMessageBox.ButtonRole.AcceptRole)
+        box.addButton("Cancel", QMessageBox.ButtonRole.RejectRole)
+        box.exec()
+        clicked = box.clickedButton()
+        if clicked is remove_all:
+            purge = True
+        elif clicked is app_only:
+            purge = False
+        else:
+            return
+        uninstall.uninstall(purge=purge)
+        QMessageBox.information(self, "RigDoctor", "Uninstalling… RigDoctor will close now.")
+        QApplication.instance().quit()
+
+    def _refresh(self) -> None:
+        self._env.setText(
+            f"Distro: {sysenv.distro_name()}    "
+            f"Package manager: {sysenv.package_manager() or 'none (apt required)'}    "
+            f"GPU: {', '.join(sysenv.gpu_vendors()) or 'unknown'}"
+        )
+        while self._components.count():
+            item = self._components.takeAt(0)
+            w = item.widget()
+            if w is not None:
+                w.deleteLater()
+
+        status = installer.component_status()
+        for component, present in status:
+            mark = "✓" if present else "✗"
+            color = GOOD if present else MUTED
+            row = QLabel(f"<span style='color:{color}'>[{mark}]</span>  "
+                         f"<b>{component.name}</b> — {component.enables}")
+            row.setTextFormat(Qt.TextFormat.RichText)
+            row.setWordWrap(True)
+            self._components.addWidget(row)
+
+        self._missing = [c for c, present in status if not present]
+        self._install_btn.setEnabled(bool(self._missing) and sysenv.package_manager() == "apt")
+        if not self._missing:
+            self._install_btn.setText("All installed ✔")
+
+    def _install(self) -> None:
+        packages = installer.missing_packages(self._missing)
+        if not packages:
+            return
+        self._install_btn.setEnabled(False)
+        self._install_btn.setText("Installing… (may prompt for password)")
+        self._output.setVisible(True)
+        self._output.setPlainText(f"Installing: {' '.join(packages)}\n")
+        threading.Thread(target=self._work, args=(packages,), daemon=True).start()
+
+    def _work(self, packages: list[str]) -> None:
+        rc, out = installer.install_packages(packages)
+        self._installed.emit(rc, out)
+
+    def _on_installed(self, rc: int, out: str) -> None:
+        self._output.setPlainText(out[-4000:])
+        self._install_btn.setText("Install missing")
+        self._refresh()
+        # If libsecret-tools was just installed, move a file token into the keyring.
+        if config.token_backend() == "file" and config.keyring_available():
+            token = config.load_token()
+            if token:
+                config.save_token(token)
+        self._refresh_update_status()
+
+    # --- update access (token) ------------------------------------------------
+    def _save_token(self) -> None:
+        token = self._token_input.text().strip()
+        if not token:
+            return
+        config.save_token(token)
+        self._token_input.clear()
+        self._refresh_update_status()
+
+    def _refresh_update_status(self) -> None:
+        self._upd_status.setText(f"{_BACKEND_DESC[config.token_backend()]} · checking…")
+        threading.Thread(target=self._check_update, daemon=True).start()
+
+    def _check_update(self) -> None:
+        self._upd_state.emit((config.token_backend(), updates.update_state()))
+
+    def _on_upd_state(self, result) -> None:
+        backend, (state, tag, _notes) = result
+        msg = {
+            updates.NO_TOKEN: "paste a token below to enable updates",
+            updates.AUTH: "token rejected — check its scope/permissions",
+            updates.NETWORK: "couldn't reach the update server",
+            updates.UP_TO_DATE: f"up to date ({tag})" if tag else "up to date",
+            updates.AVAILABLE: f"update available: {tag}",
+        }[state]
+        color = GOOD if state == updates.AVAILABLE else (WARN if state == updates.AUTH else MUTED)
+        self._upd_status.setText(
+            f"<span style='color:{MUTED}'>{_BACKEND_DESC[backend]}</span> · "
+            f"<span style='color:{color}'>{msg}</span>"
+        )

src/rigdoctor/gui/theme.py

@@ -107,4 +107,21 @@ QDoubleSpinBox, QSpinBox {{
 QTextEdit#Report {{
     background: #0d0f13; color: #cfd3da; border: 1px solid {CARD_BORDER}; border-radius: 8px;
 }}
+
+QPushButton#DangerButton {{
+    background: transparent; color: {CRIT}; border: 1px solid {CRIT};
+    border-radius: 8px; padding: 7px 14px;
+}}
+QPushButton#DangerButton:hover {{ background: {CRIT}; color: #1a0d0d; }}
+
+QPushButton#LinkButton {{
+    background: transparent; border: none; color: {MUTED};
+    text-align: left; padding: 0; text-decoration: underline;
+}}
+QPushButton#LinkButton:hover {{ color: {TEXT}; }}
+
+/* Dialogs (update prompt, changelog) — match the dark theme so text is readable. */
+QDialog {{ background: {BG}; }}
+QMessageBox {{ background: {CARD}; }}
+QDialog QLabel, QMessageBox QLabel {{ color: {TEXT}; background: transparent; }}
 """

src/rigdoctor/render.py

@@ -99,6 +99,25 @@ def _aggregate_peaks(maxima: dict) -> list[tuple[str, str, float, str, float, st
     return rows
 
 
+_SEV_LABEL = {"critical": "CRITICAL", "warning": "WARNING", "info": "INFO", "ok": "OK"}
+
+
+def render_health(findings: list) -> str:
+    if not findings:
+        return "Health report: no findings."
+    crit = sum(1 for f in findings if f.severity == "critical")
+    warn = sum(1 for f in findings if f.severity == "warning")
+    lines = ["Health report", "", f"  {crit} critical · {warn} warning · {len(findings)} checks", ""]
+    for f in findings:
+        lines.append(f"[{_SEV_LABEL.get(f.severity, '?')}] {f.category}: {f.title}")
+        if f.detail:
+            lines.append(f"    {f.detail}")
+        if f.suggestion:
+            lines.append(f"    → {f.suggestion}")
+        lines.append("")
+    return "\n".join(lines).rstrip()
+
+
 def render_summary(summary: Summary, log_path=None) -> str:
     if summary.samples == 0 and not summary.events:
         where = f" ({log_path})" if log_path else ""

tests/test_alerts.py

@@ -0,0 +1,38 @@
+"""Tests for the M8 alert monitor (edge-triggered; notify mocked)."""
+
+import unittest
+from unittest import mock
+
+from rigdoctor.core import alerts
+from rigdoctor.core.sample import Reading, Sample
+
+
+def _gpu(temp):
+    return Sample(readings=[Reading("gpu", "temp", temp, "°C")])
+
+
+class AlertTests(unittest.TestCase):
+    @mock.patch.object(alerts, "notify")
+    def test_edge_triggered_no_repeat(self, m):
+        mon = alerts.AlertMonitor(gpu_temp=90.0, cooldown=0.0)
+        mon.check(_gpu(95))   # fires
+        mon.check(_gpu(96))   # still hot — no repeat while active
+        self.assertEqual(m.call_count, 1)
+        mon.check(_gpu(50))   # clears
+        mon.check(_gpu(95))   # hot again — fires
+        self.assertEqual(m.call_count, 2)
+
+    @mock.patch.object(alerts, "notify")
+    def test_no_alert_below_threshold(self, m):
+        alerts.AlertMonitor(gpu_temp=90.0).check(_gpu(70))
+        m.assert_not_called()
+
+    @mock.patch.object(alerts, "notify")
+    def test_gpu_lost(self, m):
+        mon = alerts.AlertMonitor()
+        mon.check(Sample(readings=[Reading("gpu", "status", None, "", "query-timeout")]))
+        m.assert_called_once()
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_config.py

@@ -0,0 +1,32 @@
+"""Tests for config save/load (flat TOML writer)."""
+
+import tempfile
+import unittest
+from pathlib import Path
+from unittest import mock
+
+from rigdoctor import config
+
+
+class ConfigTests(unittest.TestCase):
+    def test_save_load_round_trip(self):
+        with tempfile.TemporaryDirectory() as d:
+            cf = Path(d) / "config.toml"
+            with mock.patch.object(config, "CONFIG_FILE", cf), mock.patch.object(config, "CONFIG_DIR", Path(d)):
+                config.save_config({"alerts_enabled": False, "gpu_temp_alert": 88.0, "update_check_minutes": 5})
+                loaded = config.load_config()
+                self.assertIs(loaded["alerts_enabled"], False)
+                self.assertEqual(loaded["gpu_temp_alert"], 88.0)
+                self.assertEqual(loaded["update_check_minutes"], 5)
+
+    def test_update_config_merges_and_keeps_defaults(self):
+        with tempfile.TemporaryDirectory() as d:
+            cf = Path(d) / "config.toml"
+            with mock.patch.object(config, "CONFIG_FILE", cf), mock.patch.object(config, "CONFIG_DIR", Path(d)):
+                config.update_config(cpu_temp_alert=70.0)
+                self.assertEqual(config.load_config()["cpu_temp_alert"], 70.0)
+                self.assertEqual(config.load_config()["gpu_temp_alert"], 90.0)  # default preserved
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_health.py

@@ -0,0 +1,46 @@
+"""Tests for the M4 health report's log scanner (synthetic input)."""
+
+import unittest
+
+from rigdoctor.core.health import CRITICAL, WARNING, run_health_checks, scan_journal_text
+
+
+class HealthScanTests(unittest.TestCase):
+    def test_xid_79_is_critical(self):
+        text = "NVRM: Xid (PCI:0000:01:00): 79, pid=1234, GPU has fallen off the bus."
+        findings = scan_journal_text(text)
+        gpu = [f for f in findings if f.category == "GPU"]
+        self.assertEqual(len(gpu), 1)
+        self.assertIn("79", gpu[0].title)
+        self.assertEqual(gpu[0].severity, CRITICAL)
+
+    def test_xid_count_aggregates(self):
+        text = "\n".join(["NVRM: Xid (PCI:0000:01:00): 79, foo"] * 3)
+        gpu = [f for f in scan_journal_text(text) if f.category == "GPU"][0]
+        self.assertIn("×3", gpu.title)
+
+    def test_oom_and_panic_detected(self):
+        text = "Out of memory: Killed process 999 (game)\nKernel panic - not syncing: x"
+        cats = {f.category for f in scan_journal_text(text)}
+        self.assertIn("Memory", cats)
+        self.assertIn("Kernel", cats)
+
+    def test_mce_critical(self):
+        findings = scan_journal_text("mce: [Hardware Error]: Machine check events logged")
+        self.assertTrue(any(f.severity == CRITICAL and f.category == "Hardware" for f in findings))
+
+    def test_clean_text_yields_no_findings(self):
+        self.assertEqual(scan_journal_text("usb 1-1: new high-speed USB device\nbluetooth: ok"), [])
+
+    def test_run_health_checks_returns_findings(self):
+        # Runs against the real system; just assert it returns a sorted list of Findings.
+        findings = run_health_checks()
+        self.assertIsInstance(findings, list)
+        severities = [f.severity for f in findings]
+        order = {"critical": 0, "warning": 1, "info": 2, "ok": 3}
+        ranks = [order.get(s, 9) for s in severities]
+        self.assertEqual(ranks, sorted(ranks))
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_installer.py

@@ -0,0 +1,46 @@
+"""Tests for the M9 installer logic and the M13 version comparison."""
+
+import unittest
+
+from rigdoctor.core import installer
+from rigdoctor.core.catalog import Component
+from rigdoctor.core.updates import is_newer
+
+
+class InstallerTests(unittest.TestCase):
+    def test_component_status_uses_presence(self):
+        status = installer.component_status(present=lambda cmd: cmd == "smartctl")
+        by_id = {c.id: ok for c, ok in status}
+        self.assertTrue(by_id["smartmontools"])
+        self.assertFalse(by_id["dmidecode"])
+
+    def test_missing_packages_dedup_preserves_order(self):
+        comps = [
+            Component("a", "A", "B", "x", ("p1", "p2"), "c1"),
+            Component("b", "B", "B", "y", ("p2", "p3"), "c2"),
+        ]
+        self.assertEqual(installer.missing_packages(comps), ["p1", "p2", "p3"])
+
+    def test_apt_command_includes_packages(self):
+        joined = " ".join(installer.apt_install_command(["smartmontools", "dmidecode"]))
+        self.assertIn("smartmontools", joined)
+        self.assertIn("dmidecode", joined)
+        self.assertIn("apt-get install", joined)
+
+    def test_install_nothing_is_noop(self):
+        rc, _ = installer.install_packages([])
+        self.assertEqual(rc, 0)
+
+
+class UpdateTests(unittest.TestCase):
+    def test_is_newer(self):
+        self.assertTrue(is_newer("v0.0.5", "0.0.4"))
+        self.assertFalse(is_newer("v0.0.4", "0.0.4"))
+        self.assertFalse(is_newer("v0.0.3", "0.0.4"))
+
+    def test_is_newer_handles_garbage(self):
+        self.assertFalse(is_newer("not-a-version", "0.0.4"))
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_inventory.py

@@ -0,0 +1,30 @@
+"""Tests for the M5 system inventory (render + dict round-trip; collect on real system)."""
+
+import unittest
+
+from rigdoctor.core import inventory
+from rigdoctor.core.inventory import Section
+
+
+class InventoryTests(unittest.TestCase):
+    def test_collect_returns_sections(self):
+        sections = inventory.collect()
+        self.assertTrue(sections)
+        titles = {s.title for s in sections}
+        self.assertIn("System", titles)
+        self.assertIn("CPU", titles)
+
+    def test_dict_round_trip(self):
+        sections = [Section("System", [("Kernel", "7.0.0"), ("Distro", "Ubuntu")])]
+        restored = inventory.from_dict(inventory.to_dict(sections))
+        self.assertEqual(restored[0].title, "System")
+        self.assertEqual(restored[0].items, [("Kernel", "7.0.0"), ("Distro", "Ubuntu")])
+
+    def test_render_markdown(self):
+        md = inventory.render_markdown([Section("CPU", [("Model", "Test CPU")])])
+        self.assertIn("## CPU", md)
+        self.assertIn("- **Model:** Test CPU", md)
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_token.py

@@ -0,0 +1,36 @@
+"""Tests for update-token storage (file fallback + env override), keyring mocked out."""
+
+import os
+import tempfile
+import unittest
+from pathlib import Path
+from unittest import mock
+
+from rigdoctor import config
+
+
+class TokenStorageTests(unittest.TestCase):
+    def test_file_fallback_roundtrip(self):
+        with tempfile.TemporaryDirectory() as d:
+            token_file = Path(d) / "token"
+            with mock.patch.object(config, "_secret_tool", return_value=None), \
+                 mock.patch.object(config, "TOKEN_FILE", token_file), \
+                 mock.patch.dict(os.environ, {}, clear=True):
+                self.assertIsNone(config.load_token())
+                config.save_token("abc123")
+                self.assertEqual(config.load_token(), "abc123")
+                self.assertEqual(config.token_backend(), "file")
+                self.assertEqual(token_file.stat().st_mode & 0o777, 0o600)
+                config.clear_token()
+                self.assertIsNone(config.load_token())
+                self.assertEqual(config.token_backend(), "none")
+
+    def test_env_override_wins(self):
+        with mock.patch.object(config, "_secret_tool", return_value=None), \
+             mock.patch.dict(os.environ, {"RIGDOCTOR_TOKEN": "envtok"}, clear=True):
+            self.assertEqual(config.load_token(), "envtok")
+            self.assertEqual(config.token_backend(), "env")
+
+
+if __name__ == "__main__":
+    unittest.main()