Add targeted web search guidance at key analysis steps: earnings call
highlights, industry context, anomaly explanation (margin shifts, debt
changes, CapEx spikes), quality-of-earnings validation, and earnings
call transcript supplementation. Includes guardrails to keep searches
targeted (3-6 total) and source-quality hierarchy.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Adds a new bundled skill for analyzing company financial statements with
a structured 9-step framework including Buffett competitive advantage
scoring, quality of earnings assessment, and SEC filing qualitative
analysis. Includes reference files for sector benchmarks and scoring
criteria.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Refactor QR code component with extracted hooks (useQRToken, useCountdown, useCopyToClipboard)
- Extract reusable sub-components (QRCodeFrame, ExpiryTimer, CopyLinkButton)
- Extract QRCodeCard and DevicesCard components in clients page
- Add GatewayStatus indicator for connection state
- Minor styling fix for empty device list state
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
The listStates IPC handler now waits for any "starting" status to
settle (max 3.5s) before returning, ensuring the UI always gets the
final status (running/error) instead of the transient "starting" state.
Also fixes unused variable lint warning in profile page.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add CompactionItem component that renders a subtle inline notification
in the chat stream when the agent compacts its context window. Extends
Message type with "system" role and handles compaction_end events in
useChat's handleStream.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add `dev:desktop:reset` to delete ~/.super-multica directory
- Add `dev:desktop:fresh` to reset and start app in one command
Enables testing the complete onboarding flow as a new user.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Rename channels.tsx to clients.tsx
- Remove standalone skills.tsx and tools.tsx pages
- Add agent/ directory for agent-related pages
- Update layout and navigation structure
- Add tabs component to ui package
- Update fonts and global styles
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Root cause: compaction only ran after successful assistant response,
never before calling the LLM API. Internal runs skipped it entirely,
and the fire-and-forget void swallowed errors with race conditions.
Layer 1 - Pre-flight compaction in transformContext:
Runs before EVERY LLM call (including internal runs). Prunes tool
results and drops oldest messages when utilization exceeds 80%.
Pure in-memory, no disk writes.
Layer 2 - Fix post-hoc compaction:
Store compaction promise instead of voiding it, await at _run() start
to prevent race conditions. Emit compaction_start before the actual
compaction. Remove useless catch-rethrow block.
Layer 3 - Context overflow recovery:
Catch context overflow 400 errors → auto-compact → retry (up to 2
attempts). Runs before classifyError/auth rotation so overflow errors
are handled by compaction, not profile switching.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- ESTIMATION_SAFETY_MARGIN: 1.2 → 1.5 (50% buffer covers CJK text)
- estimateSystemPromptTokens: /3 → /2 (conservative for mixed content)
This makes the 80% compaction trigger fire earlier, reducing the gap
between estimated and actual token counts that caused overflow errors.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Detects context overflow 400 errors from various LLM providers
(prompt too long, context length exceeded, request too large, etc.)
for use in auto-compaction recovery.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add stop button to interrupt agent generation mid-stream. The send button
toggles to a stop icon during loading. Abort propagates from UI through
IPC to the Agent layer (PiAgentCore.abort()), preserving all partial
content in the agent's context so users can follow up immediately.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Send "typing" chat action while LLM generates a response, with 5s
interval refresh and 60s safety timeout. Stop on message_end,
agent_error, or generic error events.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Gateway source moved out of src/ subdirectory but Dockerfile still
referenced old paths, breaking docker build. Rewrite to use tsx runtime
(no tsc/esbuild compile step), copy all workspace package.json for
lockfile compatibility, and filter install to gateway deps only (616 vs
1911 packages). Also update build-and-push script to tag latest.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Wrap page content with container class for consistent max-width
- Move overflow-auto to outer div so scrollbar stays at viewport edge
- Add self-stretch to vertical divider in home page
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Convert update-notification component from Hugeicons to Lucide icons
- Add UpdateNotification to layout.tsx for auto-update functionality
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add silent option to setProvider to suppress toast notifications
- Improve device list with better state handling
- Update onboarding setup step with silent provider switch
- Minor UI tweaks in layout and App components
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add AppState module in core for managing app state persistence
- Add app-state IPC handlers for reading/writing onboarding state
- Hydrate onboarding state from file system on app startup
- Prevent flash by showing blank screen during hydration
- Update onboarding store to sync with file system
- Improve MulticaIcon with enhanced animation states
- Minor UI fixes in chat and device list components
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Replace @hugeicons/react with lucide-react across all packages
- Update all components to use Lucide icon components
- Add silent option to store refresh methods to control toast display
- Simplify icon usage with direct component imports
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Scroll cursor into view on content changes (e.g., Shift+Enter)
- Add scroll threshold and margin for better edge handling
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add collapsible capabilities section showing skills, tools,
channels, and scheduled tasks counts
- Add AI brain icon to represent agent capabilities
- Add refresh button with tooltip to refresh all capabilities
- Add desktop-specific Toaster component (uses local ThemeProvider)
- Show all capability counts even when zero
- Change "View all" buttons to outline style for better distinction
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Add new stores: skills, tools, cron-jobs, hub
- Remove wrapper hooks: use-channels, use-cron-jobs, use-hub,
use-provider, use-skills, use-tools, use-heartbeat
- Update all components and pages to use stores directly
- Add 800ms minimum loading time for refresh operations
- Add toast notifications for store actions (success/error feedback)
- Remove unused chat-mode store and remote-chat component
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Add remark-math + rehype-katex plugins to render inline ($...$) and
display ($$...$$) math expressions. Includes dark mode CSS overrides,
streaming block splitting for math fences, and math range exclusion
in link preprocessing.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Web tools are always available, so the conditional branching was
unnecessary. Data section now always includes the dynamic evidence
decision guidance.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The finance-decisioner used hardcoded keyword matching and arbitrary
scoring weights to decide evidence plans — the LLM reading the
improved SKILL.md already makes this decision more intelligently.
Removes: finance-decisioner.ts, its tests, runner.ts integration
(applyFinanceResearchGuidance, saveFinanceDecisionMeta,
rebuildSystemPromptWithExtra shim), and researchDecision from
session meta.
Keeps: SKILL.md improvements, sections.ts dynamic data/web guidance.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
electron-updater is a CommonJS module — ESM named imports fail at
runtime. Switch to default import with destructuring.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The heartbeat runner used agent.write() (normal write), which persisted
both the heartbeat prompt and HEARTBEAT_OK response as regular messages
visible in the UI chat history.
Switch to runInternalForResult() — a new AsyncAgent method that runs
via runInternal() (messages marked internal: true, rolled back from
in-memory state). This hides both the heartbeat prompt and response
from the UI entirely, while still persisting to JSONL for diagnostics.
The previous commit's event-stream heartbeat ACK filter remains as a
defense-in-depth layer for edge cases.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The heartbeat runner uses agent.write() (normal write), so heartbeat
ACK responses like "HEARTBEAT_OK" were not suppressed by the internal
run filter and leaked into the desktop UI chat. The Gateway path was
already fixed (Hub has delayed-start + isHeartbeatAckEvent filtering),
but the local Desktop path through AsyncAgent had no such filtering.
Add createFilteredHandler() to AsyncAgent that buffers message_start
for assistant messages and checks subsequent events with
isHeartbeatAckEvent(). Pure heartbeat ACKs are suppressed end-to-end;
all other messages are forwarded normally.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Changed forwardAssistant from false to true in sendAnnounceDirect() so
the assistant's summary response is streamed to the desktop UI in
real-time. The announcement prompt stays internal (hidden from UI), but
the user now sees the completion notification.
Previously, persistAssistantSummary saved the response to JSONL but
never emitted events to the UI subscriber, leaving users with no
visible feedback after subagent tasks completed.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
