Picks up three releases of MatrixRTC / delayed-event fixes and
matrix-sdk-crypto-wasm 18.3.1 -> 18.9.0.
SDK 43 moved OAuth (MSC3861) into the SDK and dropped oidc-client-ts, so
the native-OAuth sign-in is ported:
- login discovers the auth server via /auth_metadata, registers with
OAuth2.registerClient, and keeps the PKCE verifier + device in its own
sessionStorage entry keyed by `state` (single use; CSRF check);
- refresh is done by the SDK (oauthClientId + onTokenRefresh); the #16
cross-tab protection is kept by a narrow guard on refresh-token grants
(the SDK calls the global fetch for OAuth, not fetchFn), which
serialises them with a Web Lock and adopts a sibling tab's tokens;
- logout and forced logout revoke via mx.logout() (the SDK revokes at
the auth server for OAuth sessions);
- the callback page parsed its params on every render, re-running the
single-use code exchange in a loop; parsed once now.
Also: notification_type is deprecated in MSC4075 (ring in DMs, notify
elsewhere when omitted), _onRTCSessionMemberUpdate ->
ensureRecalculateSessionMembers, precache limit 6 -> 10 MB for the
7.8 MB crypto wasm.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017PCb59uY3UKqDqt7B6hzd5
On the desktop app with a working keychain (Windows Credential Manager),
the tokens leave localStorage: it keeps a token-less shell blob
(tokensIn: keychain) and the full session lives in memory, read from the
keychain at boot before anything reads the session, so getFallbackSession
stays synchronous. Web, Linux and older desktop builds are unchanged.
- A new session (login, token refresh) is written to localStorage in full
first; the mirror verifies the keychain copy and only then strips the
tokens, so one durable store always holds the newest tokens.
- Existing installs migrate on first boot the same way; a failing keychain
write keeps the tokens in localStorage.
- If a shell is there but the keychain entry is missing or unreadable, the
app shows Retry / Sign in again instead of logging out, and the mirror
is not started (it would clear the entry).
Fixes#105
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017PCb59uY3UKqDqt7B6hzd5
Step 1 of moving the desktop app's login out of the webview's plaintext
localStorage: keep a verified copy of the tokens in the OS keychain
(Windows Credential Manager, via cinny-desktop's new secure_session_*
commands). The session is still read from localStorage exactly as before,
so nothing about login changes and a keychain problem can't log anyone out.
Step 2 (a later release, once this has run on real installs) switches reads
to the keychain and drops the tokens from localStorage.
- sessions.ts: onSessionPersisted — listeners told about every session
write (login, token rotation) and removal (logout); a throwing listener
can't break the write.
- keychainMirror.ts: desktop only. Mirrors userId/deviceId/accessToken/
refreshToken (not the rest of the session); reads first and writes only
when the copy differs, then verifies by reading back; serialized, 5 s
timeouts; no session → clear (also covers a logout whose reload beat the
clear). Every failure is a status, never an exception. A desktop build
without the commands reads as "unsupported", so this can ship before the
desktop side.
- Settings → General (desktop): "Login in the system keychain" status.
Tested: unit tests (fake keychain: store, no rewrite when current, rotation,
clear, unsupported, missing commands, denied write, read-back mismatch,
timeout); a simulated desktop app with a fake keychain, 14/14 (login
mirrors only the secrets, Settings status, reload verifies without
rewriting, logout clears, an existing session is mirrored after upgrade,
Linux/denied show an honest status and stay logged in); the real Linux
desktop binary (commands answer "unsupported", login unaffected, Settings
says so). Unit 1295 pass, Playwright 20 passed.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01PPmy3tPq869XDW4njjVaKA
Clears the clean 🟡 remainders from the feature audit (gate-green, 677 tests):
- F3: getFallbackSession prefers the session-blob/legacy source with the later
expiresAt (a downgrade→upgrade could boot on a stale blob's dead token).
- F6: server-forced logout (SessionLoggedOut) now mirrors logoutClient —
pushSessionToSW() + best-effort revokeOidcTokens for OIDC sessions (the search
plaintext wipe was already added).
- N5: deleteUnreadInfo parent fallback `?? roomId` → `?? []` (latently spread the
roomId string into chars).
- P10: useUserPresence re-seeds when the User object appears after first render.
- forward: strip m.mentions so forwarding doesn't re-ping the original mentions.
Left open: F5 (OIDC expiry not reachable in persistTokens), N6/H10/D7 (minor /
runtime-verify). See LOTUS_TODO.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Session now persists as ONE atomic cinny_session_v1 JSON write (blob-wins read,
transparent migration from the ~10 legacy keys, dual-write kept one release for
rollback). subscribeSessionChanges + useSessionSync reload a tab whose session
was changed/removed by another tab (logout/login/token rotation). OIDC refresher
already routes through setFallbackSession, so rotations stay atomic. Tests 7→22.
Full token-protection redesign remains tracked in LOTUS_BUGS.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
setFallbackSession gains an optional `extra` arg (password call sites unchanged)
persisting cinny_refresh_token, cinny_expires_at (absolute), and
cinny_oidc_{issuer,client_id,redirect_uri,id_token_claims}. getFallbackSession
reads them back (expiry as remaining lifetime); removeFallbackSession + re-save
clear stale OIDC keys. Session type gains `oidc?: OidcSessionMeta`. +2 tests.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Lotus CIandClaude Opus 5.5
jaredandClaude Opus 4.8
Lotus Bot
Ajay Bura