* fix(ui,server): normalize Windows drive letter casing for consistent path resolution
Provider settings were failing to persist for specific projects on Windows
because path normalization was inconsistent across the codebase. Some
normalizePath functions uppercased the Windows drive letter (c:\ -> C:\)
and others did not, causing:
- directoryScoped cache misses (different keys for the same directory)
- broken model selection in the affected project
- lost conversation history (sessions could not match their project)
- false cache hits in resolveConfigDirectory on undefined inputs
This change extracts a single shared normalizePath utility and uses it
from the 5 client sites that were missing the drive letter normalization.
The server-side normalizePathForPersistence is updated to uppercase the
drive letter both before and after safeRealpathSync, so the persisted
path is consistent even when realpath returns a symlink/junction with
a lowercase drive letter on some Windows environments.
Fixes#2109
* test(ui,server): add coverage for Windows path normalization
Address review feedback on #2154:
- Add Windows-platform test for normalizePathForPersistence covering
drive letter uppercase on input and after realpath resolution
- Add dedicated test suite for the shared normalizePath utility
- Defensive fix: normalizePath now returns null for paths that consist
only of slashes (\\, ///), matching the documented contract
Refs #2154
* fix(server): scope drive casing normalization to Windows
---------
Co-authored-by: bashrusakh <bashrusakh@users.noreply.github.com>
Co-authored-by: Bohdan Triapitsyn <artmore@protonmail.com>
Adds section headers and tooltip help for the goal settings area
Reorganizes chat settings into Session Assistance, Reasoning, Message Appearance, Tools & Files, and Composer groups
Adds localized labels and search entries for the new section headings
Relay connect used to serialize a dead LAN probe (up to 8s per stale address
on mobile, 2-4s on desktop) in front of the relay attempt, then paid a second
WebSocket connect + E2EE handshake because the probe tunnel was thrown away.
- mobile probeConnectionCandidates: race the relay probe against the direct
chain with a 1.5s direct headstart; a live LAN still wins, a dead one no
longer delays startup
- relay probes adopt their tunnel as the runtime tunnel (adoptRelayTunnel)
instead of dialing a fresh one — applies to auto-connect, pairing redeem,
password login, and the desktop host switcher's relay fallback
- relay probe drops the /health round-trip: the E2EE handshake already proves
the server identity, /auth/session alone proves liveness and auth
- desktop restoreDesktopRelayRuntime: same headstart race; a late direct
success hot-switches back (stable runtimeKey); startup probe now passes
expectedServerId so a re-leased LAN address never sees the token
- launch splash shows 'Connecting to device: <label>' with animated dots
under the (still centered) logo, translated in all locales
- editing a saved instance no longer rebuilds it from the URL field alone:
the id is passed through, relay/https candidates are preserved, and a
token-key change migrates the Keychain token instead of orphaning it
All local instances share the data dir and therefore the relay identity
(serverId), so concurrent relay hosts evicted each other at the relay worker
(4001: Control replaced) and paired devices landed on whichever local process
won last — often a stale dev server, surfacing as 'Unable to reach server'
and devices stuck on relay with 503s on newer endpoints.
- relay/host-lock.js: per-machine claim file (relay-host.lock, {pid}); stale
claims from dead pids are ignored; unwritable data dir falls back to
pre-lock behavior
- relay/service.js: start only when the claim is free or ours, otherwise
'standby' with the holder pid in lastError; 30s watcher takes over when the
claimant dies and stands down when another process claims; pairing-link
creation and explicit /relay/enable force-claim (user intent wins)
- mobileConnections.ts: log candidate-refresh skip reasons and the refresh
result instead of failing silently
Candidates refresh (server + mobile + desktop clients):
- GET /api/client-auth/connection/candidates returns the server's current
LAN URLs, relay candidate, and serverId for already-paired devices
- /health and /api/version expose serverId so clients can verify a learned
address belongs to the expected server before sending their bearer token
- mobile: refresh saved candidates over the live transport after every
connect/wake, hot-switch relay->LAN when a fresh address is reachable;
serverId gate on direct probes; token no longer sent to /health
- desktop: refresh stored host apiUrl after a relay connect and hot-switch
back to direct; electron probe verifies serverId before authenticated fetch
Fixes found while debugging a dead pairing:
- settings: strict reader that throws on corrupt/unreadable file instead of
returning {}; relay signing/encryption key generation is now gated on it,
so a swallowed read failure can no longer mint a new server identity and
orphan every paired device (loud log when a keypair IS generated)
- SessionAuthGate: bounded auto-retry for transient session-check failures
(initial request racing the relay tunnel's first WS attempt, startup 5xx)
- the goal dialog renders as the shared MobileOverlayPanel bottom sheet
on mobile instead of a centered dialog
- tapping the target button to ARM keeps the soft keyboard open (the next
message is the objective; same guard as the attachment/mic buttons),
while opening the manage sheet lets the keyboard close as usual
- capacitor: bottom-sheet overlays cap their height by the keyboard inset
AND the top safe area — with the keyboard raised while typing inside a
sheet, 100dvh does not shrink (native resize is off) and the panel could
slide under the notch/status bar; applies to every MobileOverlayPanel
The distillation moved from PlanView into setSessionGoal — the shared
choke point for every goal source (composer, fork dialog, plans) — and
into the scheduled-tasks goal creation server-side. Any objective over
5000 chars is distilled into completion criteria by the small model; on a
transient failure a head+tail excerpt keeps the intent (top) and the
acceptance criteria (bottom) with a trim marker, plus a toast. The
working agent always received the full prompt in chat — only the auditor
is bound by the limit.
Plan-goal objectives are capped at 5000 chars for the auditor while the
working agent reads the full plan from its file. Plans over the limit are
now distilled by the small model into completion criteria (end goals +
per-phase verification, no implementation steps), prefixed with a header
pointing back at the plan file so every continuation re-anchors on the
live source of truth. If distillation fails (transient small-model
hiccup), a head+tail excerpt keeps the plan's intent (top) and acceptance
criteria (bottom) with a trim marker between — sacrificing the
implementation middle the agent reads from the file anyway — and a toast
tells the user the objective is degraded.
Removes recent-project grouping from the sidebar project list
Keeps project ordering based on manual and default sort rules only
Reduces memoized session-derived state in the sidebar
Move per-session permission auto-accept policy ownership from the UI to the
OpenChamber server so enabled sessions continue running when clients disconnect
or the server restarts.
- persist explicit per-session policies in OpenChamber settings
- inherit the nearest explicit policy across subagent session hierarchies
- allow child sessions to opt out of an inherited parent policy
- immediately accept matching global and directory-scoped pending requests
- process future requests without requiring a connected UI client
- reconcile pending permissions after startup and event-stream reconnects
- deduplicate concurrent requests and retry transient reply failures
- synchronize policy updates across connected clients
- migrate existing browser-persisted policies to server storage
- suppress auto-accepted permission cards before they enter UI state
- show deduplicated permission toasts for inactive sessions
- preserve foreground-only permission handling in VS Code
- integrate directory-aware notification routing from main
- add coverage for persistence, inheritance, retries, reconciliation, pending
requests, client hydration, and inactive-session toasts
Use authoritative session.idle and session.error events for notifications
while retaining legacy message.updated handling for compatibility.
Classify sessions through targeted, directory-aware session lookups instead
of fetching the full session list. Cache confirmed root and parent session
relationships without treating failed lookups as root sessions.
Honor subagent completion settings and templates across the server-driven
web, desktop, and mobile paths, and bring the VS Code webview notification
policy to feature parity.
Use freshly synchronized VS Code settings, retry failed settings syncs,
extract session error messages, and deduplicate authoritative and legacy
completion and error events.
Adds /craft-goal autocomplete and chat handling for starting a Goal crafting session.
Introduces new Magic Prompts content and localized labels/descriptions for Goal crafting.
Migrates desktop draft starters to include Craft a Goal once and persists the migration marker.
* fix(mobile): always mount SessionSidebar to eliminate >10s drawer open delay (#1695)
On mobile (Android PWA), SessionSidebar was conditionally mounted via
{mobileLeftDrawerVisible && ...}, causing the component to unmount on
drawer close and remount on every open. Each remount fired a full
data-loading cascade: paginated sessions fetch (PAGE_SIZE=500 with
retry), worktree discovery, repo status, PR status, 10+ useMemo
recomputations, and localStorage reads, manifesting as a >10s delay
before the drawer became interactive.
Desktop already avoided this by keeping SessionSidebar always mounted
inside <Sidebar> with a CSS visibility toggle.
Fix: remove the mobileLeftDrawerVisible conditional wrapper so
SessionSidebar stays mounted on mobile too, matching desktop behavior.
Visibility remains controlled by the leftDrawerX transform (off-screen
when closed). Added pointer-events-none when hidden as a defensive guard.
Added a regression test that fails if the conditional mount pattern is
reintroduced around the mobile SessionSidebar.
* fix(mobile): hide closed drawer to avoid rotation offset leak
---------
Co-authored-by: Bohdan Triapitsyn <artmore@protonmail.com>
Compaction fixes (observed in a real long run):
- the summary message's zeroed tokens froze the goal counter at its
pre-compaction value; segments now close with the previously displayed
total as a continuity floor
- audits and continuations after a summary tail now take execution params
(provider/model/agent/variant) from the newest non-summary assistant
turn instead of inheriting agent 'compaction' and the summarize model
File-backed objectives:
- the objective text lives in <data-dir>/goals/<sessionId>.md, keyed by
session id (one goal per session, a new goal overwrites the file);
metadata carries only an objectiveFile flag so session.updated fanout
stays light, and never a path — ids are pattern-validated before any
filesystem access
- limit raised to 5000 chars, no snapshot field: the UI fetches content
via PUT/GET/DELETE /api/goals/objective/:sessionId (behind the blanket
/api auth gate), writes the file before stamping metadata, and falls
back to an inline objective when the write fails
- the loop reads the file fresh on every tick, so objectives are
live-editable mid-goal; a missing file falls back to the inline text
- scheduled goal tasks write the objective file server-side; VS Code
degrades to the audit note (route unavailable there by design)
Arm the target button in the composer and the next prompt becomes a goal:
the server keeps the session working toward it (idle tick -> small-model
audit -> continuation) until the objective is verifiably complete, blocked,
or out of budget — even with the UI closed.
Server (packages/web/server/lib/session-goal):
- event-driven loop on the global SSE hub; goal state lives in
session.metadata.openchamber.goal (merge-safe patches, stale-write guard
by goal id), so it survives restarts and syncs to every client for free
- the small-model audit (objective + last assistant turn only, language
pinned to the objective) is the sole termination authority; blocked needs
3 consecutive verdicts, audit outages tolerate one unaudited continuation
then stop the goal as resumable-blocked
- hard stops: optional token budget, auto-continuation cap (Resume grants a
fresh allowance), turn errors; user abort pauses the goal instead of
blocking it, and resuming over an aborted tail nudges immediately
- token accounting as a snapshot of the latest turn (input + cache.read +
output), goal-relative via a creation baseline and segmented across
compactions; a compaction summary skips the audit and continues
- continuations reuse the session's own provider/model/agent/variant
UI:
- three-mode target button (arm / disarm / manage dialog), informational
goal strip with inline pause/resume and an Evaluating indicator, sidebar
state glyph, objective length counter (2000-char server clamp),
read-only completed goals
- goal entry points: composer (sessions and drafts), start-new-session-
from-answer dialog, plan implement dialog (plan content becomes the
objective), scheduled tasks (Run as goal + budget)
- Settings -> Chat -> Goal: feature toggle + default token budget with
three-layer parity (web server, client persistence, VS Code bridge);
VS Code renders goal state but hides the entry points (the loop runs in
the web server only)
Notifications: per-turn "ready" notifications are suppressed while a goal
is active; settling sends one final notification (desktop, web-push, APNs
generic titles with the session name as body) honoring the completion
toggle. Error/question/permission notifications are untouched.
Docs: user guide (session-goals) in all 9 locales + sidebar entry,
scheduled-tasks cross-reference, server module DOCUMENTATION.md.
* docs: add SDK v1.17.12 migration plan — phase 4 (session.permission)
* feat(permissions): verify pending permission before auto-accept via SDK v1.17.12
Adds createPermission() and fetchPermission() wrappers on OpencodeService
for the new v2.session.permission endpoints (OpenCode SDK 1.17.12).
fetchPermission() is used by the auto-accept sweep in
resyncBlockingRequestsForDirectory to verify a permission is still
pending before replying. The auto-accept flow now skips permissions
that are already resolved, returning a null from fetchPermission()
rather than blindly calling respondToPermission on a stale entry.
createPermission() is exposed for future programmatic permission
creation; the V1 list/reply path used by the UI is unchanged.
The plan doc at plans/opencode-v1.17.12-sdk/ was rebased onto
origin/main in the prior commit to keep the PR diff focused on
this change.
Closes#1972
* fix(permissions): drop confirmed-resolved permissions from auto-accept resync
fetchPermission() now returns a tagged FetchPermissionResult so the
auto-accept loop can distinguish a server-confirmed 404 (the
permission is no longer pending) from a fetch failure (network error
or pre-v1.17.12 server). Previously both cases collapsed to null, so
a permission the server had already answered would still appear in
the resync output and trigger a spurious 'Permission needed' toast.
The auto-accept loop in resyncBlockingRequestsForDirectory now tracks
both accepted and resolved permissions, then drops both from the
'grouped' map before it falls through to the toast path. On a
pre-v1.17.12 server (no V2 endpoint) the call still returns
'unknown' and the permission stays in the resync output so the user
can answer manually — fail-closed, no false-resolved signals.
Adds a focused unit test for fetchPermission (4 cases: 200 ok, 404
resolved, 500 unknown, network throw) mocking the V2 SDK client
shape.
---------
Co-authored-by: bashrusakh <bashrusakh@users.noreply.github.com>
* perf(worktree): skip unchanged store updates and content-aware persist
- Add content-aware equality check before setState in all three discovery
loops (SessionSidebar, ElectronMiniChatApp, MobileApp). Compares
Map size and per-entry length + element references — avoids triggering
16+ subscriber re-renders when discovery finds the same worktrees.
- Add content-hash guard to persistWorktreeMap subscription with try-catch.
Avoids redundant localStorage writes when the Map reference changed but
the content is identical. Serialization errors are caught and skipped.
Contributes to #1990
* perf(worktree): extract shared worktreeMapsEqual, fix comparison, avoid double serialization
- Extract worktreeMapsEqual() into worktreeManager.ts as a shared utility
comparing worktree maps by path (not reference identity). This replaces
the inline reference-comparison logic in all three discovery loops
(SessionSidebar, ElectronMiniChatApp, MobileApp) that was ineffective
because readStableProjectWorktrees creates new object instances on
each call after cache expiry, making item !== value[i] always true.
- Pass pre-serialized JSON to persistWorktreeMap to avoid double
JSON.stringify on every persist. The subscriber already computes the
serialized string for the content-hash check; pass it through instead
of re-serializing inside persistWorktreeMap.
- Deduplicate 3 copies of the same comparison logic into the shared util.
* refactor(worktree): make worktreeMapsEqual generic over path-bearing type
The helper's equality contract is element-wise path comparison,
not anything specific to WorktreeMetadata. Generifying on
`T extends { path: string }` documents the contract at the type
level and keeps it reusable for any future map-of-arrays shape
that has a path field. Call sites stay compatible since
WorktreeMetadata has a required `path: string`.
No runtime change.
* refactor(worktree-store): clarify persist hash name and signature
Drop the optional preSerialized parameter from persistWorktreeMap —
its only caller (the subscriber) already builds the serialized
string for the content-compare, so the dual-path body is dead code.
persistWorktreeMap now takes the serialized string directly.
Rename lastPersistedWorktreeHash → lastPersistedWorktreeSerialized
(the variable holds the full JSON string, not a hash) and drop the
try/catch around JSON.stringify: it cannot realistically throw on
Map.entries() of WorktreeMetadata (no circular refs, no BigInt, no
custom toJSON). The try/catch around setItem stays — it can throw
on quota errors.
No behavior change in the success path.
* docs(worktree): trim repeated call-site comments
Replace the 5-line explanation block (copy-pasted in all three
discovery loops) with a one-liner that points at the worktreeMapsEqual
JSDoc. The '16+ subscribers' framing is also dropped — the helper
itself is general-purpose and the precise number was fuzzy.
* fix(worktree): compare branch in worktreeMapsEqual to avoid stale sidebar label
The helper compared entries by path only. An external git checkout
between discoveries changes branch (and the derived label /
headState) while path stays the same, so the helper returned true
and the store update was skipped — leaving a stale branch label in
the sidebar until the next worktree create/remove or project switch,
since there is no periodic worktree-list refresh.
Compare branch in the inner loop alongside path. Tighten the generic
constraint to T extends { path: string; branch: string } so the
contract is documented at the type level.
worktreeStatus is intentionally NOT compared: status transitions go
through setStoredWorktreeStatus, which writes a fresh Map reference
that the persist subscriber picks up directly. Adding worktreeStatus
to the contract would also force the sidebar to detect status changes
that the persist path already handles, and would couple this helper
to a field whose semantics differ from the discovery path.
Fixes the staleness concern raised by openchamber-bot in PR #1992.
* test(worktree): cover worktreeMapsEqual edge cases
Documents the helper's equality contract and guards against
regressions in the path+branch comparison. Eight cases:
- two empty maps
- identical entries (path and branch match in order)
- same path, different branch — the F1 regression case
- different paths at the same index
- per-project array length mismatch
- project-key count mismatch
- positional reorder (helper is order-sensitive)
- non-first-entry branch difference (subset detection)
All 10 tests in the file pass (2 existing + 8 new).
* ci: retrigger checks
* test(worktree): add benchmark for worktreeMapsEqual and persist path
Documents the actual cost of the PR #1992 optimizations on representative
sizes (1-1000 worktrees per project, 1-50 projects), so future contributors
can reproduce the numbers and detect regressions in the equality helper or
the persist subscriber.
Run with: `bun run packages/ui/src/lib/worktrees/worktreeManager.bench.ts`
Measured on V8 (one example run):
- worktreeMapsEqual early-exit (50×20 with first project differing):
412 ns/op vs 33,034 ns/op full sweep — ~80x speedup when any project
actually changed.
- F1 path+branch overhead vs path-only (10×50): +2.3 µs (+15.8%) on a
full sweep; on the early-exit path the F1 cost is irrelevant.
- Stringify dedup in persistWorktreeMap subscriber: 67% saved (552 µs
per persist on 10×50). This is the main absolute win of the PR.
- Content-compare guard: 19-29 ns/op, free relative to the stringify it
gates.
Bench file is standalone (import.meta.main guard) — does not run as part
of `bun test`, does not import React, does not touch localStorage.
---------
Co-authored-by: bashrusakh <bashrusakh@users.noreply.github.com>
* fix(number-input): stepper drift on rapid clicks (closes#2053)
The shared NumberInput stepper buttons (-/+) computed the next value from
`baseValue`, a useMemo of the controlled `value` prop. When the user
pressed - and + in rapid succession, both inline closures read the same
pre-update `baseValue` because the prop round-trip (click ->
onValueChange -> store/persistence update -> re-render) had not landed.
Net result: rapid alternation drifted or oscillated instead of returning
to the start value.
Route the stepper math through a new `committedValueRef` updated
synchronously inside `commitValue`, and re-sync the ref from `baseValue`
via a useEffect so external mutations (the reset button next to each
stepper, undo, multi-instance sync) keep the ref aligned. Keep
`baseValue` for the `disabled` predicate so the prop still gates the
buttons at the bounds.
For the same invariant, route `handleBlur`'s finite-parse branch
through `commitValue` so a typed value followed by a stepper click does
not compute from a stale ref. The empty-draft `onClear` early-return
relies on the baseValue useEffect to re-sync.
Cover the path with a new bun:test suite that drives the real onClick
closures through createRoot with a minimal document/window stub (no new
deps). Tests assert rapid --+ and +-- sequences net to the start value,
a sustained 6-click alternation does not drift, sequential clicks with a
re-render between them settle correctly, and a typed-then-stepper
sequence uses the typed base.
* test(number-input): restore DOM globals and guard empty recorded arrays
Follow-up to the stepper-drift fix on the same PR.
- installDomStub now captures the previous values of
document/window/navigator/IS_REACT_ACT_ENVIRONMENT before overwriting
and exposes a restore() function. withHandle calls stub.restore() in
finally after unmount(), so the test process no longer leaks a fake
DOM across tests.
- Replace the four 'recorded[length-1]!' non-null assertions with a
lastCommit(handle) helper that throws a clear error if the parent
never produced a commit. A regression that drops the first commit
fails loudly instead of silently coercing to undefined.
- Add a short comment on the useEffect re-sync documenting the
controlled-parent assumption (ref can briefly lead the prop if a
parent ever rejects or debounces onValueChange; no production caller
does today).
---------
Co-authored-by: bashrusakh <bashrusakh@users.noreply.github.com>
* feat(command-palette): add projects to existing fuzzy search
Adds projects to the existing command palette search — same single-input
fuzzy search that already covers sessions, files, settings, and commands.
Projects are scored alongside everything else by scoreByFuzzyQuery, and
the best-matching result appears first regardless of type.
Selecting a project opens a new session draft with the project pre-selected.
Closes#976
* fix(command-palette): keep file search tied to debounced query
---------
Co-authored-by: bashrusakh <bashrusakh@users.noreply.github.com>
* feat(settings): add editor font size setting for chat input and code editor
Adds an 'Editor font size' control in Settings > Appearance that sets an
absolute px font size for the chat input textarea and the in-app
CodeMirror editor. Mirrors the existing terminalFontSize lifecycle.
- New store field editorFontSize (default 13, clamp 9-32, step 1) in
useUIStore with narrow selectors at each consumer.
- Persistence wired through appearanceAutoSave, desktop + runtime API
types, and persistence.ts read/normalize.
- Settings UI row (NumberInput) with reset to 13, VisibleSetting union
entry, OpenChamberPage registration, and search index entry
appearance.editor-font-size.
- Applied as a post-zoom absolute override on the chat input textarea
and on the CodeMirror theme's content rule, leaving gutter/line-number
chrome at its existing hardcoded sizes (matches terminal scope).
- All 10 locales translated (en, es, fr, ja, ko, pl, pt-BR, uk, zh-CN,
zh-TW); no English placeholders in non-English dictionaries.
Refs #1325
* fix(codemirror): use unitless lineHeight so it scales with editor font size
The & rule in the CodeMirror theme set lineHeight to 1.5rem (~24px),
which does not scale when editorFontSize is increased (e.g., 28-32px).
This causes overlapping lines at larger font sizes.
Change to unitless 1.5, which scales proportionally with whatever fontSize
resolves to (dynamic prop or --text-code fallback). Matches browser best
practice for proportional leading.
Review comment: https://github.com/openchamber/openchamber/pull/2065
---------
Co-authored-by: bashrusakh <bashrusakh@users.noreply.github.com>
Co-authored-by: Bohdan Triapitsyn <artmore@protonmail.com>
Greg Haynes