Security
- Default scripts_management, callbacks_management, links_management, and
media_folders_management to False so a leaked token cannot escalate to RCE
through admin CRUD endpoints.
- TokenSpec + scope hierarchy (read | control | admin); legacy bare-string
api_tokens entries promote to admin for back-compat. Management endpoints
now require admin scope.
- WebSocket subprotocol auth (Sec-WebSocket-Protocol: media-server.token.<T>)
preferred over ?token= query so the token no longer lands in URL/history/
Referer; query fallback retained for HA integration back-compat.
- Origin allow-list check on the WS endpoint (CSWSH defence).
- In-process token-bucket rate limiter: 5/min for failed auths,
10/min for /api/scripts/execute and /api/callbacks/execute.
- shell=False subprocess path (shlex.split) + per-parameter regex `pattern`
in ScriptParameterConfig to harden shell=true scripts against parameter
injection (Windows cmd.exe env-var expansion).
- CSP gains form-action, worker-src, manifest-src directives.
- Refuse cors_origins=["*"] at startup; strip token=... from uvicorn access
logs; validate Gitea release tag against strict SemVer regex.
- noopener noreferrer + no-referrer referrerpolicy on every outbound link.
- icacls hardening of config.yaml on Windows (current user + SYSTEM +
Administrators only); 0600 still enforced on POSIX.
- WS volume handler clamps input and never drops the socket on bad messages.
Performance
- Album-art read in windows_media gated by track key — was decoding the
WinRT thumbnail twice per second regardless of track changes.
- /api/media/artwork returns content-derived ETag + Cache-Control so the
browser sends If-None-Match and gets 304s on track repeats.
- Foreground-service ctypes argtypes hoisted to one-time module init
(was re-declaring ~14 prototypes per probe).
- display_service _static_cache keyed by (edid_hash, ...) tuple with
eviction of disappeared monitors — fixes stale capabilities on hot-plug
swaps where the new topology has the same monitor count.
- Visualizer rAF loop paused on document.hidden, resumed on visible.
Reliability / bug fixes
- Lifespan rewritten as try/yield/finally so a partial-startup failure
cannot orphan background tasks or executors.
- _run_callback in routes/media.py keeps a strong task ref (GC-safe) and
uses the dedicated callback executor instead of the default pool.
- macos_media.set_volume() no longer always returns True.
- TrayManager._restart_requested initialised in __init__; set before
signalling exit so the main thread observes it correctly.
- Missing static_dir now logs a WARNING instead of silent UI disable.
UX / accessibility / PWA
- manifest.json theme_color and background_color match the Studio Reference
base (#0E0D0B); added id and scope for PWA installability.
- ARIA on mini-player icon buttons; inner SVGs marked aria-hidden.
- OS mediaSession API wired so headset / lockscreen / Bluetooth buttons
drive play/pause/next/prev/seek and show track metadata + artwork.
Observability
- X-Request-ID middleware (accept upstream id if it matches a safe regex,
otherwise UUID4); request_id_var added to ContextVars and included in
every log line alongside the token label.
- Audit log (append-only JSONL) for every script + callback execution,
including the on_play/on_pause/etc. event callbacks. Background-thread
writer; queue capped; flushed in lifespan teardown.
Deployment
- proxy_headers + forwarded_allow_ips plumbed through Settings →
uvicorn.Config for reverse-proxy installs.
- HTTPS support via ssl_certfile + ssl_keyfile (+ optional password);
startup refuses to launch with only one of the pair set.
- Thumbnail cache moved from project-root .cache to
%LOCALAPPDATA%/media-server/cache (Windows) and
$XDG_CACHE_HOME/media-server/thumbnails (POSIX).
Tests
- 35 new tests across auth scopes, rate limiter, browser path traversal
(../ NUL UNC absolute), script-param validation incl. regex, Gitea tag
whitelist, config atomic write + POSIX perms. 47 passed / 4 skipped.
Adds cross-platform foreground-window tracking and exposes it over REST
(/api/foreground) and the existing WebSocket feed.
- foreground_service.py: Windows probe via ctypes (HANDLE-correct argtypes
to avoid 64-bit handle truncation); macOS via AppKit; Linux via Xlib
(Wayland returns unavailable). TTL cache + per-platform fallback.
- browser_url_service.py: when foreground is a recognised browser, extract
the page title from the window title (browser-name suffix stripped) and
surface `is_browser` + `browser_page_title`. Optional UIA-based URL
extraction behind MEDIA_SERVER_BROWSER_UIA env flag (off by default —
Chromium browsers keep their accessibility tree dormant otherwise).
- websocket_manager: poll foreground every 1s inside the existing status
loop, broadcast `foreground` on connect and `foreground_update` on
change. Diff only on user-visible fields to avoid geometry spam.
- WebUI: new editorial card rendered under the monitor list on the
Display tab — process name, window title, fullscreen/minimized/monitor
chips, browser block when applicable, exe path, PID, started-ago,
geometry, platform. 16px inter-section gap matches Settings cadence.
- i18n: 25 new keys added to both en.json and ru.json.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Security
- Default bind 127.0.0.1; first-run bootstrap generates random api_token
and refuses to bind non-loopback without auth unless explicitly opted in
- Path-traversal hardened: BrowserService.validate_path rejects absolute
paths, drive letters, UNC, NUL bytes. /api/browser/{play,metadata,
thumbnail} now require folder_id and a folder-relative path
- Pydantic validators on links: http(s) URLs only, mdi:<slug> icons only
- Scripts/callbacks/links create/update/delete gated by *_management flags
- Strict CSP, X-Frame-Options DENY, Referrer-Policy no-referrer,
X-Content-Type-Options nosniff
- CORS locked to localhost:<port> + 127.0.0.1:<port> by default; configurable
- config.yaml writes atomic (tmp + os.replace) and 0o600 on POSIX
- Subprocesses spawned in their own process group / new session so timeout
kills the whole tree (Windows CREATE_NEW_PROCESS_GROUP, POSIX
start_new_session=True)
- Frontend XSS: monitor name + details escapeHtml'd; power button moved to
delegated data-action handler; remote MDI SVGs parsed and sanitized
(strip script/foreignObject/on*/javascript: hrefs) before innerHTML
- All dynamic URL segments now wrapped in encodeURIComponent
Bugs
- WebSocket reconnect: close previous socket before opening new, clear
ping interval per-socket, clear reconnectTimeout up-front, retry on
online/visibilitychange, try/catch JSON.parse
- Artwork fetch race: AbortController + generation guard
- _broadcast_after_open: initialize status, swallow per-poll errors,
background tasks tracked in a strong-ref set with done-callback cleanup
- Audio analyzer: sticky _unavailable flag prevents infinite start/stop
spin when no loopback device exists; cleared by set_device()
- Volume short-circuit cache invalidated when server reports remote volume
- Browser thumbnail race: per-folder generation counter + isConnected
checks; aborts in-flight fetches on navigation
- Track-skip uses cached title instead of full WinRT status round-trip
Performance
- Linux MPRIS/pactl and /api/display DDC-CI handlers wrapped in
asyncio.to_thread so blocking IO never stalls the event loop
- browse_directory moved off the event loop (SMB shares could freeze it)
- Windows status poll caches one asyncio loop per worker thread via
threading.local instead of new_event_loop/close on every 0.5s tick
- broadcast() serializes JSON once and uses send_text to all clients
- Hourly thumbnail cache cleanup scheduled in lifespan (was never invoked
— cache grew unbounded)
- Progress drag listeners attached only while dragging
Quality
- All asyncio.get_event_loop() in coroutines → get_running_loop()
- ThreadPoolExecutors shut down cleanly during lifespan teardown
- config_manager dedup: 12 near-identical methods collapsed onto generic
_upsert/_delete helpers (~290 lines removed)
- Service worker no longer pass-throughs every fetch
- M3U playlist written via NamedTemporaryFile (no fixed-path symlink
clobber race)
- __version__ now prefers live pyproject.toml in dev checkouts so
pip install -e . users see the source-of-truth version, not the stale
package-metadata version baked in at install time
UI/UX (Studio Reference)
- Green leftover focus rings (rgba(29,185,84,...)) all replaced with
copper accent (rgba(var(--copper-rgb),...))
- Dialogs: square corners, copper top hairline, unified with editorial
chrome
- .browser-item: transparent with copper hover border (was filled card)
- Audio device select uses var(--sans) instead of generic system font
- Mobile container padding tuned for ≤480px screens
- Breadcrumb home is a real <button> with aria-label; aria-current on root
- i18n: filled display.msg.power_*, execution.*, scripts.params.execute,
callbacks.empty in both en + ru
Frontend hot path (player.js, background.js):
- visualizer rAF: drop per-frame getComputedStyle('--accent') (cached on
applyAccentColor), build canvas LinearGradient once per accent change
instead of 32× per frame, batch all bars into a single beginPath/fill
- FPS-gate canvas redraw via frequencyDataVersion so 60-144 Hz monitors
stop re-rendering identical frames produced at 30 Hz on the backend
- editorial spectrum bars: replace style.height (layout) with
transform: scaleY (compositor-only); cache bar refs, pre-compute
per-bar gain/range, dedup writes at 1/1000 quantization
- coalesce VU needle into the visualizer rAF; cache vuNeedle ref;
dedup angle writes at 0.1°
- updateUI: status-payload fingerprint short-circuits the redundant
status_update broadcasts that fire during a track change
- swapArtworkSrc: only force layout reflow when keyframe is in flight;
drop the ?_=Date.now() cache-buster so identical artwork URLs reuse
the decoded bitmap; mini/glow imgs only re-set src when changed
- drop the fullscreen MutationObserver — fs-bloom-art is mirrored
directly from the artwork-swap path, eliminating the second blur paint
- updateProgress: skip text writes when the rounded second hasn't moved;
POSITION_INTERPOLATION_MS 100 → 250
- background.js: lift resizeBackgroundCanvas out of the rAF body, cache
step, accept new int-scaled wire format
CSS:
- spectrum bars use transform: scaleY(var(--bar-h-scale)) + transition
on transform; will-change updated to transform
- album-art-glow and fs-bloom-art switched to small-source-blur trick
(render at 20-25% size, scale 4-6×, lower blur radius) — visually
equivalent, ~10-25× cheaper repaint on track change
- drop unused transition: filter on .vinyl-stage #album-art
Backend (audio_analyzer.py, websocket_manager.py):
- pre-allocate windowed and cumsum buffers; replace
np.concatenate(([0.0], np.cumsum(...))) with cumsum[0]=0 +
np.cumsum(out=cumsum[1:]); float32 hanning window
- RMS via np.dot(mono, mono) — no astype copy, no ** temp
- int16 wire format (scale=1000) — smaller JSON, no Python float boxing
- versioned data + threading.Event so _audio_broadcast_loop is event-
driven (ev.wait + monotonic seq dedup) instead of polling on a timer
with the always-false `data is _last_data` identity check
ruff clean, pytest 7 passed / 3 numpy-skipped, esbuild bundle 113.6 kB.
Instead of waiting for the next poll cycle, new clients now get the
current playback status immediately on connect by calling get_status_func
if no cached status is available yet.
- Audio capture starts only when first client subscribes,
stops when last client unsubscribes (saves CPU/battery)
- Add lifecycle lock to AudioAnalyzer for thread-safe start/stop
- Status badge uses local visualizer state instead of server flag
- Fix script name vertical text break on narrow screens
- Fix script grid minimum column width on small viewports
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Fix CORS: set allow_credentials=False (token auth, not cookies)
- Add threading.Lock for position cache thread safety
- Add shutdown_executor() for clean ThreadPoolExecutor cleanup
- Dedicated ThreadPoolExecutors for script/callback execution
- Fix Mutagen file handle leaks with try/finally close
- Reduce idle WebSocket polling (0.5s → 2.0s when no clients)
- Add :focus-visible styles for playback control buttons
- Add aria-label to icon-only header buttons
- Dynamic album art alt text for screen readers
- Persist MDI icon cache to localStorage
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- New audio_analyzer service: loopback capture via soundcard + numpy FFT
- Real-time spectrogram bars below album art with accent color gradient
- Album art and vinyl pulse to bass energy beats
- WebSocket subscriber pattern for opt-in audio data streaming
- Audio device selection in Settings tab with auto-detect fallback
- Optimized FFT pipeline: vectorized cumsum bin grouping, pre-serialized JSON broadcast
- Visualizer config: enabled/fps/bins/device in config.yaml
- Optional deps: soundcard + numpy (graceful degradation if missing)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add LinkConfig model and links field to settings
- Add CRUD API endpoints for links (list/create/update/delete)
- Add Links management tab in WebUI with add/edit/delete dialogs
- Add live icon preview in Link and Script dialog forms
- Show MDI icons inline in Quick Actions cards, Scripts table, Links table
- Add broadcast_links_changed WebSocket event for live updates
- Add EN/RU translations for all links management strings
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Features:
- Runtime script CRUD operations (create, update, delete)
- Thread-safe ConfigManager for YAML updates
- WebSocket notifications for script changes
- Web UI script management interface with full CRUD
- Home Assistant auto-reload on script changes
- Client-side position interpolation for smooth playback
- Include command field in script list API response
Technical improvements:
- Added broadcast_scripts_changed() to WebSocket manager
- Enhanced HA integration to handle scripts_changed messages
- Implemented smooth position updates in Web UI (100ms interval)
- Thread-safe configuration updates with file locking
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
FastAPI REST API server for controlling system-wide media playback
on Windows, Linux, macOS, and Android.
Features:
- Play/Pause/Stop/Next/Previous track controls
- Volume control and mute
- Seek within tracks
- Current track info (title, artist, album, artwork)
- WebSocket real-time status updates
- Script execution API
- Token-based authentication
- Cross-platform support
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
alexei.dolgolyov