fix: memory consolidation timeout and hidden files in chat loading#1469
Open
jecruz wants to merge 33 commits intoagent0ai:developmentfrom
Open
fix: memory consolidation timeout and hidden files in chat loading#1469jecruz wants to merge 33 commits intoagent0ai:developmentfrom
jecruz wants to merge 33 commits intoagent0ai:developmentfrom
Conversation
enable websocket and api caching params
commit e48ee68 Author: keyboardstaff <keyboardstaff@gmail.com> Date: Sun Mar 29 06:27:50 2026 -0700 fix(_model_config): restore utility preset provider fallback and empty option commit c16134e Author: keyboardstaff <keyboardstaff@gmail.com> Date: Sun Mar 29 00:28:49 2026 -0700 fix(_model_config): prevent Settings Save from clobbering API keys saved via Configure Models commit 9ff4133 Author: keyboardstaff <keyboardstaff@gmail.com> Date: Sat Mar 28 10:11:57 2026 -0700 refactor(_model_config): extract reusable model-field component, split store into mixins, unify API key lifecycle commit 14de2ab Merge: 69e1774 eef6d6d Author: Wabifocus <keyboardstaff@gmail.com> Date: Sat Mar 28 19:41:43 2026 -0700 Merge pull request agent0ai#48 from agent0ai/development Development
Promote the a0_small prompt work into the shipped default stack. Token count for default prompt, accounting for tools, plugins, secrets placeholders, inactive project, etc: 2996 tokens. - migrate compact prompt fragments into core prompts, plugin-owned prompts, and the agent0 overlay - remove the obsolete a0_small profile and rename the tool-call knowledge reference to a generic framework file - keep a small set of high-value JSON examples while preserving key guardrails for tool names, tool args, and method-style calls - bake balanced conciseness into default response behavior without changing the initial greeting - drop the call_sub prompt-side loader and keep call_sub guidance self-contained - restore essential secrets/security guidance, preserve inactive-project messaging, and rebalance the A2A prompt - add a regression test that verifies the assembled default agent0 prompt stays under the token budget and retains critical guardrails
commit 5193ef7 Author: frdel <38891707+frdel@users.noreply.github.com> Date: Tue Mar 31 09:47:02 2026 +0200 refactor: change default mode from dedicated to self-chat and reorder UI settings - Change default mode to self-chat across all modules - Update README to reflect self-chat as primary mode with security warning - Move session/media storage from usr/whatsapp to tmp/whatsapp - Reorder config UI: move Mode above Allowed Numbers - Add warning banner when allowed_numbers is empty in self-chat mode - Move Bridge Port and Poll Interval to bottom of settings - Update mode descriptions to clarify self-chat handles both self commit 9fece91 Author: frdel <38891707+frdel@users.noreply.github.com> Date: Tue Mar 31 09:20:35 2026 +0200 refactor: centralize WhatsApp storage paths and improve bridge dependency handling - Add storage_paths.py helper for consistent session/media/runtime paths - Replace hardcoded usr/whatsapp paths across all modules - Fix bridge lock to be event-loop-aware (recreate per loop) - Add automatic dependency reinstall on startup failures - Track bridge startup output for better error diagnostics - Add dependency state tracking with package.json hash validation - Implement force reinstall when node_modules appears commit bc511d2 Author: linuztx <linuztx@gmail.com> Date: Tue Mar 31 09:07:46 2026 +0800 fix: stop poll loop immediately when Node.js is not installed commit a9554e1 Author: linuztx <linuztx@gmail.com> Date: Tue Mar 31 08:49:15 2026 +0800 fix: auto-reinstall corrupt node_modules and stop poll loop after repeated bridge failures _ensure_npm_install now verifies key package exists, not just node_modules dir. Wipes and reinstalls if corrupt. Poll loop stops after 5 consecutive bridge start failures instead of spamming errors and making A0 unusable. commit 61fa1bf Author: linuztx <linuztx@gmail.com> Date: Tue Mar 31 08:38:51 2026 +0800 fix: move allowed_numbers filtering from JS bridge to Python handler The JS bridge used LIDs (internal WhatsApp identifiers) for sender matching which never matched actual phone numbers. Moved filtering to Python handler.py where config is read fresh each poll cycle. - Add senderNumber (resolved phone) to bridge message payload - Filter in poll_messages() with normalized number comparison - Remove --allowed-numbers CLI arg and JS-side filtering - Fix ensure_bridge_http_up not recording _bridge_config - Fix falsy empty-dict check in bridge restart detection commit 64ee177 Author: linuztx <linuztx@gmail.com> Date: Sat Mar 28 23:34:23 2026 +0800 refactor: move email agent instructions to system prompt and update prompt labels commit 0f53b41 Author: linuztx <linuztx@gmail.com> Date: Sat Mar 28 10:59:44 2026 +0800 Add node_modules to gitignore commit eb6a4d3 Author: linuztx <linuztx@gmail.com> Date: Sat Mar 28 10:53:59 2026 +0800 Add WhatsApp plugin thumbnail commit 39bed4f Author: linuztx <linuztx@gmail.com> Date: Sat Mar 28 10:51:47 2026 +0800 refactor: rename allowed_users to allowed_numbers across plugin commit e4991b6 Author: linuztx <linuztx@gmail.com> Date: Fri Mar 27 21:58:29 2026 +0800 improve: move agent instructions from per-message to system prompt commit 4f1be15 Author: linuztx <linuztx@gmail.com> Date: Fri Mar 27 21:00:25 2026 +0800 improve: add macOS port kill support and bridge process destructor cleanup commit f534975 Author: linuztx <linuztx@gmail.com> Date: Fri Mar 27 17:09:56 2026 +0800 improve: remove redundant bridge_manager from execute, rely on poll loop finally commit 9d9dd4b Author: linuztx <linuztx@gmail.com> Date: Fri Mar 27 14:41:14 2026 +0800 fix: stop bridge and poll loop when plugin is disabled or toggled off commit 66b0a7d Author: linuztx <linuztx@gmail.com> Date: Fri Mar 27 11:05:58 2026 +0800 improve: fix allowed users input, auto-strip + prefix, log ignored messages commit 938e7b9 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 23:26:42 2026 +0800 improve: add line break to allowed users description commit 4ef64b9 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 22:44:55 2026 +0800 feat: convert markdown to WhatsApp formatting before sending replies commit f549b49 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 22:34:56 2026 +0800 improve: add progress update instructions to system context prompt commit 66e5d51 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 22:23:32 2026 +0800 fix: stop typing indicator on agent error or generation failure commit 3dd01cd Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 18:31:38 2026 +0800 improve: persistent typing indicator with poll-based refresh commit 8d0ec86 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 17:11:25 2026 +0800 Update README.md commit e664673 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 16:05:44 2026 +0800 feat: add agent prefix to self-chat replies for visual distinction commit 18c5716 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 15:43:01 2026 +0800 fix: clear typing indicator after sending reply in self-chat mode commit 7c653c9 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 14:43:06 2026 +0800 improve: merge WhatsApp Link and Disconnect into single Account field commit 57c95e6 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 14:11:05 2026 +0800 feat: add disconnect account option to switch WhatsApp accounts commit c626953 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 14:00:00 2026 +0800 improve: move mode description inline and reorder Allow Group after Allowed Users commit 18a56ea Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 13:44:17 2026 +0800 fix: remove duplicate typing indicator before sending reply commit 44c90a1 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 13:30:06 2026 +0800 improve: remove sender number from DM prompt commit 64fe7d0 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 13:17:29 2026 +0800 fix: handle documentWithCaptionMessage wrapper for captioned documents commit 00b6657 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 13:06:40 2026 +0800 feat: add attachment reader/writer with RFC and download all media types commit 8041c08 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 11:45:17 2026 +0800 improve: update group prompt and reply instructions commit 71a6eb7 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 11:26:36 2026 +0800 feat: reply to specific messages in group chats with quote commit 6bf63eb Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 09:57:34 2026 +0800 feat: detect replies to bot messages in group chats commit b4492e0 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 09:20:27 2026 +0800 improve: resolve group names and sender LIDs in bridge messages commit 14e673f Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 04:44:50 2026 +0800 feat: add allow_group toggle to respond only when mentioned in group chats commit 40f4884 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 03:20:02 2026 +0800 refactor: rename mode value from bot to dedicated commit 50af7c2 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 02:34:51 2026 +0800 fix: kill orphaned bridge process on port before starting new one commit 45b21c0 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 02:07:45 2026 +0800 improve: auto-restart bridge when config changes commit a12183b Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 01:39:55 2026 +0800 feat: add bot and self-chat mode selection for WhatsApp bridge commit bb8961a Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 00:56:56 2026 +0800 improve: send typing indicator immediately on message receive commit 84c12b0 Author: linuztx <linuztx@gmail.com> Date: Thu Mar 26 00:29:04 2026 +0800 feat: add WhatsApp integration plugin with Baileys bridge and QR pairing
…cements - Extract UI server setup into UiServerRuntime class with modular initialization - Move environment configuration, route registration, and transport handlers to helpers/ui_server.py - Add released_at timestamp tracking for git tags and branch heads across update system - Implement get_current_major_main_latest_info to find latest same-major version on main branch - Add major_upgrade_versions and main_branch_latest fields to update info payload - Remove
Add the always-enabled `_discovery` plugin to turn the welcome screen into a discovery surface for the Plugin Hub and A0 integrations. Includes a hero card plus Telegram, Email, and WhatsApp feature cards, with persistent dismiss/restore state, CTA routing to plugin config screens, and self-contained placeholder artwork. Implemented entirely through the existing WebUI extension mechanism with no core welcome-screen changes. stores cleanup layout polish and onboarding integration Move feature card titles beside thumbnails for better space efficiency and visibility. Restructure card markup and styles to support a fluid grid layout and horizontal alignment. Integrate discovery cards into the final onboarding step via a new 'onboarding-success-end' extension point, ensuring new users see extension opportunities immediately after setup. Hide discovery cards on the dashboard while the missing API key onboarding banner is visible to reduce UI noise and user confusion during initial config. update discovery card initialization and loading logic Enhance the discovery store to fetch cards from the API, improving the dynamic loading of discovery cards based on user context. This change optimizes the user experience by ensuring relevant cards are displayed immediately after onboarding and when modals are closed. And on top of that, there's a proper backend for these new cards.
quick update flow and migration guidance
`find_plugin_dir` can return `None` if a plugin cannot be found. Passing this null value to `files.get_abs_path` caused crashes during config retrieval. `get_plugin_config` and `get_default_plugin_config` now check for a valid directory and return early if it is missing.
fix: guard against missing plugin directory in config loads
feat: add plugin discovery to dashboard and onboarding wizard
prompts: adopt compact defaults and rebalance guidance
Commit 54362bf went too far stripping too many JSON examples to guide LLMs.
prompts: restore tool examples for better model guidance
Restore main.communication, main.solving, main.tips and tool.response, which made the model dumber. For some reasons the drawback was more visible with frontier LLMs. restore builtin plugins and agent0 profile
prompts: restore legacy, plugins, agent0 profile
Trying to steer the model into treating the final curly brace of a JSON obj as EOS token. Also defaulting to concise responses, expanding only when needed.
Track parsing depth via _pop_stack() helper. Exposes a 'completed' flag that signals when the root JSON structure is fully closed, allowing stream consumers to break early instead of waiting for irrelevant tokens.
Tool execution no longer waits for the full streamed assistant text. We now detect the first explicitly closed top-level JSON object, freeze that snapshot as the canonical tool request, and stop the model stream there for dispatch. To make that safe, DirtyJson completion semantics are tightened so completed=true only means the root object was explicitly closed, not that parsing hit end of file. I also restricted the new extraction path to object roots only, since tool calls are always brace-delimited objects, and added tests for parser completion and early stream stop.
Dispatch tool calls at first completed JSON object
add thumbnail for _discovery plugin
* feat: add per-row extension points inside sidebar chat list x-for loop Adds sidebar-chat-item-start and sidebar-chat-item-end x-extension points inside the x-for loop in chats-list.html. Previously only sidebar-chats-list-start/end existed, both outside the x-for loop. This forced plugins that need per-chat-row UI (e.g. status indicators, labels, badges) to resort to MutationObserver + index-based DOM scanning and monkey-patching internal store methods. With these new extension points, plugins can inject content into each chat row with access to the reactive Alpine context object (context.id, context.name, context.running, context.project, etc.) entirely through declarative Alpine bindings — no DOM scanning, no method patching, no index arithmetic. * feat: add resolve_mcp_server_headers hook + @extensible on set_settings (local PR-B+C shape) Adds resolve_mcp_server_headers async extension point at both MCP transport paths in mcp_handler.py (streamablehttp + sse). Enables plugins to resolve credential placeholders at header construction time without monkey-patching. Adds @extensible to set_settings() and set_settings_delta() in settings.py. Enables plugins to intercept settings writes for credential scanning. Local patch shape for upstream PR-B+C submission. Ref: deimos_openbao_secrets IMPLEMENTATION_PLAN.md Step 1 * fix: increase memory consolidation timeout to 300 seconds for local models * fix: ignore .DS_Store and hidden files when loading chats --------- Co-authored-by: Tristan (Deimos AI) <tristan@deimosai.com> Co-authored-by: Deimos-Agent <deimos-agent@localhost> Co-authored-by: Jan Tomášek <38891707+frdel@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Two bug fixes:
fix: increase memory consolidation timeout to 300 seconds for local models
plugins/_memory/helpers/memory_consolidation.py: changesprocessing_timeout_secondsfrom 60 to 300fix: ignore .DS_Store and hidden files when loading chats
helpers/persist_chat.py: filters out.DS_Storeand hidden filesFiles Changed
plugins/_memory/helpers/memory_consolidation.py(+2/-2)helpers/persist_chat.py(+2/-2)Checklist
developmentbranch