Use a dark scrim for MobileOverlayPanel instead of a translucent background so
it dims the rest of the screen like the sessions sheet.
Rework MobileSurfaceShell to mirror that structure: the scrim is the fixed
parent's background with a quick fade (keeping it on a compositing layer iOS
Safari clips to the viewport), and the sheet is a flex child sized to leave the
top safe area dimmed — dropping the old top shadow that pooled at the corners.
Mount Files/Changes/Settings only while open so each recomputes its
safe-area layout fresh, fixing intermittent top-inset dimming, and drop the
dim behind the header overflow menu to avoid a scrim-to-scrim flicker on
select.
Nest child sessions (by parentID) under their parent in the mobile sessions
list, with a chevron in the row's left gutter to expand/collapse them
recursively. Top-level pagination counts only parent sessions; children whose
parent isn't in the same bucket stay top-level so nothing is hidden.
Expansion state lives in an in-memory store that survives closing and
reopening the sheet but resets on a full page reload.
Restructure the changes list so the staged/changed file list scrolls inside
ChangesPanel while the commit message and actions stay pinned to the bottom,
matching the desktop git sidebar instead of letting the commit box scroll out
of view with the list.
Remove the per-row agent color dot (and its now-unused helpers), keeping only
the status indicator. Add a "+" button to the panel header that opens a new
session draft targeting the filtered project (or the most recently active
session's project under "All", falling back to the default target), and
restyle the close button to match it.
Show the context token usage next to the percentage (e.g. 5.1% 20.4K/400.0K)
at a larger, consistent size, and compute the percentage with a fraction
instead of the store's integer-rounded value.
Add a two-step archive affordance to each mobile session row, mirroring the
project remove confirmation: tapping the archive icon arms an inline confirm
(with cancel), and confirming archives the session. Only one row can be in
the confirming state at a time.
Also relabel the project remove confirmation from "Delete" to "Close"
across all locales, since removing a project closes it from the app rather
than deleting anything.
Replace the up/down arrows in the mobile project reorder list (drag already
covers reordering) with an edit button that opens a dedicated project editor
surface: rename, pick icon and color, and discover a favicon (no upload).
The editor lists the project's worktrees with drag-to-reorder (persisted per
project, like project order) and a delete button that opens a mobile
confirmation built on the shared worktree primitives — archiving attached
sessions and optionally removing the local/remote branch.
Persist project/worktree expand state across restarts and decouple it from
the active directory — projects default expanded, worktree groups collapsed,
and the user's explicit toggles are remembered.
Selecting a session now switches the active project (not just the draft),
following the session's directory.
Replace all-or-nothing session pagination with incremental +7 loading that
resets when a group or project is toggled, matching the desktop sidebar.
Always render root (project-level) sessions as a flat list above worktree
groups, and vertically center the session row status dot.
Add a toggle for displaying archived sessions in the sidebar menu
Load more sessions in smaller increments
Reset expanded session counts when collapsing groups or projects
* feat: add dialog for "Start new session from this answer"
Replace the one-click fork action on assistant messages with a dialog
(ForkSessionDialog) that lets the user pick model, thinking level, and
agent, plus edit the instructions sent to the new session.
The instructions field is prefilled with the previous fixed fork prompt
and is mandatory. The composed message is now fully visible (no synthetic
preface): the user's instructions sit above a short fixed connective that
opens the assistant content. createSessionFromAssistantMessage takes the
chosen execution params instead of reading from config.
Also fix TodoSendDialog visuals: narrower vertical layout, model trigger
no longer stretches with centered text, and the agent/thinking dropdowns
portal to body so opening them no longer nudges the dialog height. Extract
the shared ThinkingPill into its own component.
* fix: address review feedback on fork session dialog
- Fix "bellow" -> "below" typo in the fork content preface (now user-visible
since the message is no longer synthetic)
- Reset ForkSessionDialog state only on open transition, reading the config
store snapshot via getState() so background store refreshes can't discard
in-progress instruction edits
Add a packaged-client runtime boundary so the shared UI can talk to local,
desktop, remote, and VS Code runtimes through the right transport instead of
assuming one same-origin web server.
Centralize OpenChamber-owned API access behind RuntimeAPIs, runtimeFetch, and
runtime URL helpers, while keeping official OpenCode traffic on the SDK path.
Support runtime switching, remote host selection, desktop client credentials,
and headless connection links for pairing packaged clients with remote
OpenChamber servers.
Harden the new auth model by moving long-lived client tokens out of browser
URLs, introducing short-lived scoped URL tokens for browser-owned transports,
restricting URL-token access to explicit readable/realtime routes, and making
client-token management session-scoped or self-scoped as appropriate.
Update browser-owned assets and preview proxy flows to work with the split
runtime model, including authenticated project icons, preview token propagation,
CSP-safe preview bridge injection, and preview proxy auth that survives
short-lived URL-token expiry.
Tighten Electron security boundaries for packaged clients by gating privileged
preload state to trusted origins and requiring explicit confirmation before
connect deep-links import or switch remote runtimes.
Also refresh agent guidance and project skills so future runtime/API, auth,
preview, UI, CLI, settings, locale, and drag-to-reorder work follows the new
architecture.
Loading screens: drop the internal-jargon progress text on the opencode
reload overlay (ConfigUpdateOverlay) and the vscode init splash — show
text only on errors. Replace it with a glow pulse on the OpenCode mark on
the cube's top face; OpenChamberLogo's isAnimated prop was a no-op and now
actually animates. vscode shows the glow on the inline splash logo and the
React app stops writing 'Loading data (… Providers, … Agents)…'.
Restart API: 'OpenChamber: Restart API Connection' now runs the same full
reload flow used after an OpenCode update — the command asks the chat
webview to call reloadOpenCodeConfiguration() (overlay + managed restart
via the bridge + config/data refresh) instead of a bare manager restart,
falling back to the old restart when no webview is open. Mounts
ConfigUpdateOverlay in the vscode app so the overlay actually shows there.
On refresh the auto-open draft opens in an effect (after first paint), so
ChatContainer briefly rendered ChatEmptyState (logo + 'start a new chat')
before the welcome screen. When a draft is about to auto-open and there's
no init error, render a neutral background instead — the empty state stays
for the cases where it's actually meaningful (no auto-open, or init error).
Capture the @dnd-kit setup that works for both desktop and mobile —
the Translate-not-Transform stretch fix, rectSortingStrategy for wrapping
rows, MouseSensor + long-press TouchSensor, touch-action none, and the
pitfalls we hit (onDragOver live-reorder infinite loop, offset overlay).
References DraftPresetChips and sortable-tabs-strip.
Let users curate the draft welcome chips: pin existing commands and skills
as starters, remove them, and drag to reorder — all inline on the draft
screen via a '+' picker dialog and per-chip remove, with no separate
settings UI.
A starter references a command or skill; its scope is inherited from the
item (user-scope -> global, project-scope -> per-project). Global starters
persist to settings.json (useUIStore + client/server sanitizers); project
starters persist to the project config alongside worktree setup commands.
The two scopes form ordered namespaces shown global-first then project,
reorderable only within each group.
The six built-in Session magic-prompt commands are the default global set
and stay available in the picker for re-pinning if removed; they keep their
bespoke icons, while user commands/skills fall back to the Commands/Skills
section icons. Chip labels are normalized (/simplify-code -> 'Simplify
code'). Missing commands/skills are skipped rather than shown broken.
Drag-to-reorder works on desktop and mobile: rectSortingStrategy for the
wrapping multi-row layout, CSS.Translate (no scale) so the lifted chip
doesn't stretch, and MouseSensor + long-press TouchSensor so taps still
submit and swipes still scroll. The '+' picker is a searchable dialog on
every surface.
Mini-chat uses the same centered-welcome compact layout, so the preset chips fit fine under the welcome message there as well. Drop the mini-chat exclusion.
On narrow surfaces (mobile, vscode) the composer sits at the bottom and
only the welcome message is centered, so the preset chips had nowhere
sensible to live and were desktop-only. Render them under the centered
welcome message there instead.
Extract the shared preset list (draftPresets.ts) and chip row
(DraftPresetChips) so both layouts reuse them. Since the command-aware
submit lives in ChatInput, ChatContainer triggers it through a new
input-store channel (requestPresetSubmit / consumePendingPresetSubmit)
that ChatInput consumes. Mini-chat stays excluded — too small for the row.
The old workspace-review prompt only hunted for catastrophic bugs and
omitted everything it wasn't highly certain about, so it usually returned
'No high-signal issues found' and felt useless.
Rework it to first understand the diff's intent and judge whether the
implementation actually achieves it and is adequate (smallest correct
solution, completeness), then review correctness across concrete failure
modes (concurrency, lifecycle, state fanout, comparators, a11y,
regressions, targeted-test gaps) plus a security/supply-chain lens, and
classify findings by severity (blocker / non-blocker / nit). Keeps the
existing false-positive discipline (diff is source of truth, validation
pass, no nitpicks, cite rules) and stays review-only by default.
Adapted from a battle-tested PR-review prompt, minus the GitHub-specific
parts that don't apply to a local workspace diff. Updates command and
settings descriptions across locales.
Reword the /catch-up prompt and its Magic Prompts settings descriptions to
emphasize an easy-to-digest summary rather than a 'warm'/'friendly' tone —
the goal is ease of comprehension, not friendliness. Updated across all
locales.
Rework the /catch-up decision tree from mutually-exclusive cases into
complementary layers. It now determines whether the current branch is the
repo's default branch and builds context accordingly: on a feature branch
it reads enough of the branch's own commits to grasp the intent and checks
the branch's PR even when there are uncommitted changes (so in-progress
work is understood as part of the larger effort); on the default branch it
lightly skims recent commits. Uncommitted changes remain the focal point
but are interpreted through that context.
The suggested next step now favors continuing the actual work over
housekeeping — since running catch-up implies the user stepped away and
the work is likely unfinished, push/PR/checks are only a fallback. Updates
the Magic Prompts settings descriptions across locales to match.
Promote the last plain-prompt draft chip into a proper /explore command so
all draft welcome presets are now commands. Its hidden prompt investigates
the repository and gives a structured orientation — big picture, main
modules, how they connect, conventions, and where to start — instead of a
file-by-file dump.
Follows the established pattern: visible + hidden magic prompts wired into
command autocomplete, the submit handler, the draft chip, and the Magic
Prompts settings page, with i18n across all locales. Drops the now-unused
explore prompt string.
Rework the /weigh hidden prompt so the recommendation is driven by what
best serves the user's actual goal — not by what is fastest or least
effort. Effort and complexity are surfaced as honest consequences rather
than reasons to steer toward a weaker option, strong-but-harder
approaches that fully cover the idea must be included and discussed, and
a watered-down solution is never recommended just because the proper one
is more work.
Add /weigh as a draft chip and command for the moment before planning —
when you know what to build but not how. Its hidden prompt investigates
the code, then lays out 2-3 genuinely distinct approaches with trade-offs
(complexity, risk, blast radius, effort) and a clear recommendation,
without writing a plan or code.
Follows the established command pattern: visible + hidden magic prompts
wired into command autocomplete, the submit handler, draft preset chips,
and the Magic Prompts settings page, with i18n across all locales. Adds a
scales-3 icon to the sprite.
Rework the /catch-up hidden instructions: write like a teammate catching
you up rather than a CI summary, scope strictly to the current branch,
and never surface unrelated branches, other people's PRs, or review
requests. When the current branch has no PR of its own, skip pull-request
talk entirely and just recap recent work plus a next step.
Replace the 'What changed recently' draft chip with /catch-up, a command
whose hidden prompt branches on git state: reconstruct intent from an
in-progress diff, check an open PR's review state, or summarize recent
commits.
Add /debug as a new draft chip and command: a guided root-cause
investigation that captures the symptom, forms hypotheses, checks them
against the code, and confirms the cause before proposing a fix.
Both follow the /workspace-review pattern — visible + hidden magic
prompts wired into command autocomplete, the submit handler, draft preset
chips, and the Magic Prompts settings page, with i18n across all locales.
Add starter preset chips under the composer on the desktop draft welcome
screen (explore, what changed, plan, review). Clicking a chip submits
immediately; prompt chips send a natural-language prompt, command chips
reuse built-in slash commands.
Add a new /plan-feature built-in command, modeled on /workspace-review:
visible + hidden magic prompts that run a guided, batched-question
planning dialogue (investigate the code, ask up to 3 clarifying
questions at a time, surface pitfalls, then produce an implementation
plan). Wired into command autocomplete, the submit handler, the draft
plan chip, and the Magic Prompts settings page, with i18n across all
locales.
Capture the preview/browser iframe DOM with snapDOM (html-to-image
fallback) so web annotation screenshots match the visible viewport,
without a headless Chromium dependency.
- Preserve document scroll via viewport crop and re-bake nested scroll
(e.g. the Starlight sidebar) deterministically on the clone
- Pin position:fixed elements to their measured viewport rect so headers
and sidebars land correctly in the crop
- Extract preview capture/proxy helpers into
lib/preview/screenshot-capture.ts to slim down ContextPanel
- Guard the external preview proxy against SSRF to private, loopback and
reserved/link-local addresses (incl. cloud metadata)
- Fully validate preview bridge messages before formatting/use
- Warn on the empty browser tab that pages run with full access, so
users browse untrusted sites knowingly
Fix chat history pagination and scroll preservation
Align session history loading with the expected scroll-up pagination UX while
keeping OpenChamber-specific initial message limits for constrained runtimes.
- Separate initial load sizes from older-history pagination size
- Load older messages automatically when scrolling near the top
- Continue fetching history until a visible older turn is available
- Preserve the current viewport synchronously during prepends
- Prevent history loading from fighting pinned-to-bottom follow behavior
- Remove delayed scroll-to-bottom correction that caused jumpbacks
- Fix the virtualizer fallback path that could render a large blank spacer
- Track oldest loaded message per pagination iteration to avoid redundant fetches
Only auto-send queued messages after an active session becomes idle
Avoid treating missing initial status as safe to send
Add coverage for queue auto-send status transitions
Queued messages now match the reverted message dock style
Added per-message edit and send actions
Sending one queued message no longer sends the full queue
Prevents the context panel from overflowing past the right edge
Clamps resize width to the available layout space
Handles previously saved oversized panel widths
Add launch-at-startup support across the Electron desktop app and the web CLI.
Electron now supports macOS launch-at-login through the native login item API. Login launches start OpenChamber in the background without opening a window, while Dock activation, deep links, and second-instance launches still open or focus the normal app window. The desktop Settings UI now exposes a localized launch-at-login toggle in Desktop Network Access.
The web CLI now includes `openchamber startup status|enable|disable`, backed by native user services:
- macOS: launchd LaunchAgent
- Linux: systemd --user service
- Windows: Task Scheduler
Startup services run `openchamber serve --foreground` so the OS service manager owns process lifetime and restarts. Foreground service updates now defer restarts to the service manager instead of spawning duplicate CLI restarts.
Startup services snapshot useful environment variables by default so provider tokens, PATH, SSH agent settings, and OpenCode configuration survive login/reboot starts. The snapshot avoids shell/session-only state, uses systemd-compatible env quoting on Linux, and avoids unused env artifacts on macOS.
Also adds localized docs for startup services and environment variables.
Safely falls back to raw patch text for invalid diffs
Protects individual tool parts from crashing the chat
Covers malformed apply_patch and edit diff cases
Use browser keyboard resizing by default on mobile
Remove mobile-only autocomplete button and tabs
Keep autocomplete behavior consistent across mobile and desktop
Treat the mobile web surface as a constrained runtime so sync loads smaller message pages, keeps fewer warm session caches, and evicts heavy inactive sessions instead of retaining them across switches.
Limit mobile message-record and turn-model caches to reduce memory pressure on phones while preserving bounded initial page expansion for large final turns.
Split the mobile session status bar so the collapsed state avoids subscribing to the full session/status list; the expensive grouping work now only mounts for the expanded list.
Verified with bun run type-check and bun run lint.
Prevents queued messages from being sent to a newly opened session
Adds explicit session targeting for queued auto-send
Covers the behavior with a unit test
Reconnects and resyncs active sessions when live updates stall
Normalizes synthetic session status events
Uses authoritative status snapshots to clear stale busy states
* feat(chat): live markdown source-mode highlighting in composer
Highlight markdown syntax, fenced code blocks, and mention-style tokens
directly in the chat input via the existing transparent-textarea overlay
(color/decoration/background only, so caret alignment is preserved).
- Markdown source-mode: inline/fenced code, links, headings, blockquotes,
list markers, with dimmed syntax punctuation
- Per-language syntax highlighting inside fenced blocks, reusing the editor's
CodeMirror language resolver + Lezer (bash/js/ts/json/html/css/python/md);
highlighted blocks use a neutral base, plain fences keep the code color
- Token highlighting on match: @file, @agent, /command, /skill, #snippet
- Auto-pairing: wrap selection with markers, triple-backtick expands to a
fenced block; paste a URL over a selection to form a markdown link
- Add md/markdown to the shared code-block language resolver
* fix(chat): address composer highlight review
- Tilde (~~~) fenced blocks now get per-language syntax highlighting
- Share fence open/close detection between tokenizeMarkdown and
highlightFencedCode so they agree on boundaries (fence length + format),
fixing range bleed with 4-backtick fences and ```lang lines inside blocks
- Replace buildHighlightParts O(segments x ranges) scan with a sweep-line
over an active set (verified equivalent vs the prior algorithm across 30k
randomized cases, including overlaps and explicit class/priority)
- Cap per-block Lezer parsing at 20k chars; oversized blocks keep the neutral
code base without per-token coloring
* docs: add OpenChamber feature docs and translations
Add 30 new docs pages covering OpenChamber-specific workflows and setup:
OpenCode server, providers/models/agents, MCP, skills, commands & snippets,
usage, projects, context, notes/todos/plans, scheduled tasks, project actions,
preview, worktrees, multi-run, git & GitHub, magic prompts, git identities,
mobile/PWA, security, notifications, voice, project icons, remote instances,
desktop browser, updates, and three troubleshooting pages.
Rebuild sidebar into eight task-oriented sections and translate every new
page into all six supported locales (uk, zh-cn, es, pt-br, ko, pl).
* docs: surface new sections on homepage and cross-link tunnels
Add an Explore block to the docs homepage (all seven locales) linking to
the new section anchors, and cross-link the Tunnels page to Security and
PWA & Mobile.
* perf(server): cache deterministic git rev-parse reads in fs exec route
A fresh client (e.g. immediately after a page reload) has an empty git
store and re-resolves every project's root from scratch, firing identical
`git rev-parse --absolute-git-dir` / `--git-common-dir` lookups against
`/api/fs/exec`. Each spawns a git subprocess server-side, so re-opening a
workspace recomputes everything.
Add a small TTL cache for an allowlist of deterministic, side-effect-free
git plumbing path queries, keyed by `(resolvedCwd, command)`:
- Only `git rev-parse` path lookups (absolute-git-dir, git-common-dir,
show-toplevel) are cacheable; any other command — including any non-git
command — always executes and is never stored.
- Only successful results are cached (failures may be transient).
- TTL is configurable via OPENCHAMBER_GIT_READ_CACHE_TTL_MS (default 30s,
0 disables). The git directory layout is effectively static while the
app runs, so a short TTL safely absorbs the post-reload burst.
- Expired entries are pruned alongside exec jobs.
Complements the client-side root-resolution cache: that one collapses the
in-session N² cascade, this one absorbs the cold-start burst on reload.
Adds tests covering cache hit, per-cwd keying, non-allowlisted commands,
failed-result bypass and the disable switch.
* fix(server): bound git-read cache with count + byte limits; test TTL expiry
Per the project caching policy (AGENTS.md: cap in-memory caches with both
count and byte limits), the git-read cache was unbounded between prunes.
Add dual-constraint LRU eviction (500 entries / 1MB, oldest-first) with
recency refresh on cache hits. Add tests for TTL expiry (fake timers) and
count-cap eviction.
* fix(server): dedupe in-flight git read cache hits
* perf(git): cache project-root resolution to stop N² polling cascade
Opening a workspace with many projects/worktrees fired hundreds of
`POST /api/fs/exec` requests (e.g. ~700 for 19 projects) within seconds,
dominated by repeated `git rev-parse --absolute-git-dir` /
`--git-common-dir` for the same directories.
Root cause: in `useProjectRepoStatus`, each project's `ensureStatus`
settles independently and mutates the git store, which re-derives
`projectGitBranchesKey` and re-runs `getRootBranch` for *all* projects on
every change. `getRootBranch` had no caching, so this produced an N×N
burst of uncached git plumbing calls.
Changes:
- worktreeStatus: extract `resolveProjectRoot` to module scope with a
60s TTL cache + in-flight dedupe (root resolution is static within a
session). Combine the two `rev-parse` queries into one subprocess.
Add `getRootBranch(dir, { knownBranch })` fast-path that skips a
redundant git status when the directory is its own root, while still
resolving the primary-root branch correctly for linked worktrees.
Export `invalidateResolvedProjectRootCache`.
- useProjectRepoStatus: replace the cascade effect with a debounced,
diff-based pass that only resolves projects that are new or whose
branch actually changed, passing the known branch through.
- worktreeManager: invalidate the root cache on worktree create/remove.
- Add unit tests for caching, dedupe, invalidation, rev-parse
precedence, non-git fallback, linked-worktree resolution and the
knownBranch fast-path.
Reduces startup from hundreds of requests to roughly one root
resolution per project.
* fix(git): clear in-flight resolves and guard write-back on cache invalidation
`invalidateResolvedProjectRootCache` cleared `resolvedRootCache` but left
`inFlightRootResolves` intact, so during a worktree topology change a
resolution already in flight could (1) be handed to callers arriving after
invalidation and (2) re-seed the cache with the pre-invalidation root when it
settled, defeating invalidation for up to the full TTL.
Drop the in-flight entry on invalidation and add an epoch guard so a resolve
that was invalidated mid-flight does not write its now-stale result back.
Add a regression test for the concurrent-invalidation scenario.
* fix(git): bound root cache and avoid early sidebar resolves
Add a Voice & Style section to the docs authoring guide, then bring every
docs page in line with it: lead with the task, add success signals to
procedures, explain jargon on first use, keep bullet casing consistent,
and link out to Troubleshooting where steps can fail.
Applied across English source and all localized versions (uk, zh-cn, es,
pt-br, ko, pl).
Transfers Electron app as a tarball between release jobs
Verifies the macOS app executable before packaging
Documents permission risk in Tauri migration flow
Keep the left sidebar open when the context panel opens
Reduce right sidebar and context panel default widths
Collapse only the sync button label at medium widths
Replace the prompt-template workflow with snippet support that is compatible with opencode snippet conventions. Snippets are now stored and loaded from global and project snippet directories, including legacy pluralized paths, with frontmatter metadata for aliases and descriptions. Snippet expansion supports recursive references plus prepend and append sections, while inject sections are treated as unsupported no-ops so OpenChamber remains compatible without requiring an external plugin.
Add the snippets settings experience and remove the old prompt-template settings surface. The new settings page and sidebar support creating, editing, deleting, selecting, and describing snippets, with localized copy across every supported locale. The settings navigation now exposes Snippets with a dedicated icon and metadata.
Wire snippets into all prompt-entry surfaces that need them. Chat, multi-run groups, and scheduled task prompts now offer hash-trigger snippet autocomplete and expand snippets before sending work to OpenCode. Chat also uses an adaptive compact placeholder on mobile or narrow composer widths so helper trigger guidance stays readable in constrained layouts.
Keep multi-run aligned with grouped prompts. Multi-run sessions now use a shared title builder that handles both legacy titles and the newer g1, g2 prompt-group title format. Fusion parsing now recognizes grouped multi-run titles, scopes fusion sources to the same prompt group, and creates fusion sessions under the matching group so outputs from different prompts are not mixed accidentally.
Harden the icon sprite pipeline. The sprite generator now discovers icon names used through typed icon maps, JSX icon props, IconName returns, and generated-value flows without scanning unrelated string literals or the generated sprite itself. The generated sprite is strictly typed so invalid icon names are caught by type checking, and existing invalid or unsafe icon references were cleaned up across settings, provider, Git identity, scheduled task, voice, header, and sidebar surfaces.
Update backend configuration routes and documentation for snippets. The OpenCode config route layer now exposes snippet CRUD and expansion endpoints, accepts JSON bodies for snippet writes, and removes the old prompt-template provider. Scheduled task runtime expansion now uses snippets before dispatching messages.
Add regression coverage for snippet storage and expansion, config-route JSON handling, and multi-run title parsing. Validated with full type checking, full linting, targeted multi-run title tests, and targeted OpenCode snippet/config route tests.
Repackages Electron app for legacy Tauri updater migration
Stops release and manual DMG workflows from building Tauri
Documents transition flow and cleanup timing
Improve chat session switching and history pagination, with most of the aggressive limits scoped to the VS Code webview where the freezes were observed.
Session history loading and pagination:
- Reduce the VS Code message page size to 30 records so switching sessions does not immediately hydrate large histories into the webview.
- Keep manual Load older messages in VS Code fixed at 30 records per request instead of growing the request size over time.
- Add a bounded VS Code initial-tail expansion path from 30 to 50, 80, and 120 records only when the initial page has no user-message turn boundary, preventing large final turns from rendering as an empty chat.
- Lower the normal web message page size from 200 to 150 for a mild shared optimization without adopting the aggressive VS Code limits.
- Make session pagination metadata reactive per session so ChatContainer receives cursor updates from materialization and reconnect paths without requiring a switch away and back.
- Write pagination metadata before publishing newly materialized messages so the first render sees the correct has-more state.
- Store cursor information from direct materialization and reconnect message fetches in the shared session prefetch metadata cache.
VS Code cache and memory pressure reductions:
- Use a shared per-directory session recency map so cache eviction is based on app-level recency instead of whichever useSync instance happened to run.
- Limit VS Code warm session cache retention to 4 sessions and evict heavy inactive message caches after switching away from a large session.
- Disable sidebar session prefetch in VS Code because warming extra sessions was increasing webview memory and GC pressure during navigation.
- Remove dropdown background message prefetch so opening the switcher does not start additional session materialization work.
- Drop cached session-message-record snapshots when evicting session data so stale derived records do not remain after the raw session cache is cleared.
- Add bounded LRU caching for session message record snapshots, with much smaller VS Code limits and a VS Code cap that avoids caching snapshots above 30 messages.
- Bound the turn-window model cache in VS Code and avoid caching turn models for sessions above the VS Code message-page size.
Chat render-path reductions:
- Reuse ChatContainer's already-materialized message records in plan detection instead of adding a second active-session message subscription.
- Add a no-op guard when marking session plan availability so repeated detections do not create new Map references and fan out renders.
- Add no-op guards for session switcher and dropdown open state updates to avoid unnecessary store updates and renders.
- Convert several session-specific hooks to useSyncExternalStore with empty-session no-subscribe behavior so empty IDs do not subscribe to broad store updates.
- Remount the chat viewport when the current session changes, isolating per-session viewport and list state.
- Change the virtualized message-list fallback to render only a tail window when the virtualizer has not produced rows yet, instead of rendering an entire large history.
VS Code layout and header improvements:
- Remove the broad useSessions subscription from the VS Code layout header path and subscribe only to the active session title and initial-session existence.
- Unmount the compact VS Code session sidebar when the user is in chat view instead of keeping the hidden session list mounted and subscribed.
- Compute the latest assistant model and latest context-token usage in a single reverse scan of current-session messages instead of scanning the same list twice.
- Remove switcher git-status warmup work so the switcher reads already-loaded branch labels without starting extra background git status requests.
Markdown and file-reference safeguards:
- Skip expensive syntax highlighting for very large code blocks, with a 200-line cap in VS Code and a softer 1200-line cap in web.
- Add an LRU cap to file-reference stat lookups so the cache cannot grow without bound across many rendered messages.
- Limit the number of file references annotated per render to 40 in VS Code and 200 in web to prevent large assistant outputs from spawning too many stat checks.
- Clear file-link annotations when file-reference mode is disabled so stale attributes and handlers do not remain on previously annotated nodes.
Assistant-message action and preview reductions:
- Skip preview URL scanning on VS Code, mobile, and mini-chat surfaces so assistant text and tool output are not scanned where the preview action is unavailable.
- Skip Save-as-Plan project lookup on VS Code, mini-chat, and mobile surfaces.
- Hide Save-as-Plan and Start MultiRun assistant-message actions on VS Code, mini-chat, and mobile surfaces.
- Resolve the current session directory on demand for assistant actions instead of subscribing each assistant message to the full session list.
Tool and task rendering optimizations:
- Prefer finalized task metadata summaries without fetching child-session messages when the summary is already present.
- Avoid polling or final-fetching task child sessions once a final metadata summary is available.
- Use VS Code-specific task child fetch limits of 30 records for initial, active, and idle fetches.
- Parse diff stats by scanning patch text line-by-line instead of splitting large patches into arrays.
- Count write-tool lines by scanning content instead of allocating a split array for large files.
- Avoid trimming large patch strings just to test whether they contain content.
- Memoize diff and write statistics so unchanged tool parts do not recalculate them on every render.
VS Code bridge improvements:
- Return JSON and text proxy responses through the VS Code bridge as bodyText instead of base64 so the webview avoids synchronous base64 decoding for common API responses.
- Keep binary responses on the base64 path while making bodyBase64 optional in the bridge contract.
- Strip content-length, content-encoding, and transfer-encoding headers from proxied responses because the bridge reconstructs the Response body.
Validation:
- bun run type-check
- bun run lint
- bun run vscode:build
Bohdan Triapitsyn