Prefer markdown responses from servers that support Cloudflare's Markdown
for Agents feature, reducing token usage by ~80% when available. Non-supporting
servers fall back to HTML as before.
- Update Accept header to prefer text/markdown in web_fetch requests
- Add text/markdown content-type handling to skip HTML parsing pipeline
- Capture x-markdown-tokens response header in WebFetchResult
- Add extractMarkdownTitle() helper for native markdown title extraction
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add optional context parameter to getAuthHeaders() so callers can
provide feature-specific suffixes (e.g. "to use web search") in the
not-logged-in error message, restoring the original behavior.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace duplicated getLocalAuth() + manual header construction in
finance/api.ts and web-search.ts with the shared getAuthHeaders()
and API_BASE_URL from hub/api-client.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Extract API_BASE_URL and getAuthHeaders() into a reusable module
so that tools don't duplicate base URL and auth header construction.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Route all financial data requests through api-dev.copilothub.ai/api/v1/financial
proxy and authenticate via sid/device-id/os-type headers instead of X-API-KEY.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Migrate web_search tool from HMAC-SHA256 reqId signing to
sid/device-id/os-type auth headers, matching the desktop API client
pattern. Update endpoint to /api/v1/web-search.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Reads sid and deviceId from ~/.super-multica/auth.json for use by
tools that need authenticated API access.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add common generateEncryptedId() utility in @multica/utils
- All Device IDs now use same encryption algorithm (40 hex chars)
- Web: store encrypted format directly in localStorage
- Desktop: use shared utility, accept encrypted ID from Web
- Hub: use shared utility for hub-id generation
- Telegram: use shared utility for device ID generation
- Gateway hook: use encrypted format for client connections
Algorithm: sha256(sha256(uuid).slice(0,32)).slice(0,8) + sha256(uuid).slice(0,32)
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Wraps the read tool from pi-coding-agent to automatically downscale
oversized images (>1MB or >2000px) before they enter the session.
Uses macOS sips for resize with no extra dependencies, following the
same pattern as OpenClaw. Falls back gracefully on non-macOS.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Session JSONL files were bloated because base64 image data was stored
inline (a real session had 6.7MB of images in a 9.8MB file). Images
are now extracted to per-session media/ directories as binary files,
with compact $ref references stored in the JSONL. Images are restored
transparently on read. Old sessions with inline base64 remain
backward compatible and auto-migrate on next compaction.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Codex OAuth tokens use the ChatGPT backend API (chatgpt.com), not the
standard OpenAI API (api.openai.com). pi-ai already has a dedicated
openai-codex provider with the correct API format (openai-codex-responses)
and base URL. Remove the alias that was incorrectly mapping openai-codex
to openai, which caused 401 errors due to missing scopes.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Adds a status bar and slide-out dashboard panel to the chat view that
shows subagent progress in real-time. Polls at 2s when active, 10s when
idle. Completed runs auto-hide after 5 minutes. Includes dismiss button
and 30s auto-dismiss for the status bar.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- detectSplitTurn now returns adjustedToSummarize (truncated at
assistantIndex) so split prefix messages are not summarized twice
and removedCount is not inflated
- Remove unused contextWindow parameter from computeAdaptiveChunkRatio
- Remove duplicated single-chunk fast path in compactMessagesWithChunkedSummary
(the multi-chunk loop handles 1 chunk naturally)
- Fix triple-newline formatting in buildPlainTextFallback by concatenating
metadata sections directly instead of joining through parts array
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Default compaction mode changed from "tokens" to "summary" (LLM-based)
- Split turn detection: handles orphaned tool_result messages at compaction
boundary by summarizing the split turn prefix separately
- Adaptive chunk sizing: dynamically adjusts chunk ratio (0.15-0.4) based on
average message token count instead of fixed 50k tokens
- Fallback chain: summary compaction wraps in try/catch, falling back to
tokens mode on complete failure; summarizeWithFallback provides 3-level
degradation within summary mode itself
- Metadata appended to summaries: file operations + tool failures sections
- CompactionResult extended with fileOperations and toolFailures fields
- CompactionEndEvent gains optional summary field
- API key resolution improved: reuses agent's own key for summary generation
Backward compatible: explicit "count" or "tokens" mode behaves identically.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Codex and Claude Code OAuth tokens with refresh_token can auto-renew,
so they should not be considered expired based on last_refresh time or
file mtime. Previously, credentials were hardcoded to expire after 1
hour, causing valid Codex sessions to show as unconfigured.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Move channel plugin initialization (initChannels + ChannelManager
constructor) before restoreAgents() to prevent TypeError when
createAgent() calls channelManager.listChannelInfos() during restore.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Create onChannelSendFile callback in Hub.createAgent() that tries the
channel plugin path first (local file), then falls back to the gateway
path (base64 over RoutedMessage). Also pass channel info to agent.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add ChannelInfo type and buildChannelsSection() that informs the LLM
about connected messaging channels and their capabilities (e.g. send
files). Wire through SystemPromptOptions and runner.ts.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Prepend [ChannelName · private/group] prefix to debounced messages so
the LLM knows the message source. Add sendFile() for outbound media
routing and listChannelInfos() for system prompt channel awareness.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
New send_file tool with TypeBox schema, auto-detect media type from
file extension, and file validation. Wired through AgentOptions and
resolveTools with conditional registration when callback is provided.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add OutboundMedia interface and OutboundMediaType to the channel type
system. Implement sendMedia in the Telegram plugin using grammy's
InputFile API with HTML caption formatting and plain-text fallback.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Guide the LLM to evaluate snippet quality after web_search and
follow up with web_fetch on the most relevant URLs when deeper
content is needed for accurate answers.
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Hidden directory ~/.super-multica is not user-friendly for a working
directory. Move default workspace base to ~/Documents/Multica so
users can easily find agent-created files in Finder.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add MessageSource type to track where user messages originate (local, gateway, channel)
- Broadcast inbound messages from all channels to local Desktop UI via Hub.onInboundMessage()
- Persist source field in JSONL session storage so it survives page refresh
- Display source icon (Monitor/Smartphone/Send) with tooltip for non-local user messages
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Each agent profile now gets a dedicated workspace directory
(~/.super-multica/workspace/{profileId}) used as the default CWD
for tool operations. Supports override via MULTICA_WORKSPACE_DIR
env var or config.json workspaceDir field. The workspace path is
injected into system prompt and runtime info.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
A subagent that times out loses all its work, so a generous default
reduces wasted compute. Update the constant, tool description, system
prompt guidelines, and documentation to reflect the new 1800s default.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
When writeInternal fails with a transient error (e.g. undici
"terminated", connection reset, 502/503), wait 5 seconds and retry
once. This prevents continuation tasks from silently failing when
the LLM streaming connection is interrupted mid-generation.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
sessions_spawn now accepts groupId and next parameters. First spawn with
next auto-creates a group; subsequent spawns join via groupId. sessions_list
groups runs by groupId with completion progress display.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
When all runs in a group complete, deliver combined findings plus the
`next` continuation prompt to the parent agent via writeInternal. The
parent can then act on the collected data (summarize, write files, etc).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add SubagentGroup for "collect all, then act" workflows where multiple
subagents complete before a continuation task runs. Groups are persisted
alongside run records and support a `next` continuation prompt.
Co-Authored-By: Claude Opus 4.6 <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>
- 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>
