Add user-configurable timezone and 12/24-hour preferences, then wire them through settings, runtime snapshots, scheduler payloads, wait handling, notifications, backups, memory, plugin metadata, and frontend formatters.
Keep UTC as the boundary for absolute instants while serializing user-facing dates in the configured or browser-resolved timezone. Preserve scheduler wall-clock inputs in the selected timezone, propagate TZ into desktop/runtime process environments, and restart active desktop sessions when the runtime timezone changes.
Cover the risky paths with timezone regression tests for settings normalization, auto and fixed timezone resolution, scheduler round-trips, memory timestamp conversion, and desktop timezone sync.
Detect stalled automatic history compression so the prompt-prep wait loop cannot spin forever when no further reduction is possible.
Split large manual chat compaction input by verified token budget instead of line midpoint, covering single-line 85k+ character histories.
Add regression tests for stalled compression, max-pass bailout, and large single-line compaction chunking.
Validate and persist API chat lifetime: lifetime_hours is validated as a positive number and stored in the AgentContext data, and context.last_message is set using UTC. Removed the in-class threading-based cleanup state and old _cleanup_expired_chats method. Introduced a new job-loop extension (extensions/python/job_loop/_20_cleanup_expired_api_chats.py) that periodically scans AgentContext instances and removes expired API chats (using persist_chat.remove_chat) in a UTC-aware manner. Added tests (tests/test_api_chat_lifetime.py) to verify lifetime persistence and that the job loop removes expired chats.
Standardize multi-action tools around tool_args.action while keeping parser compatibility for older tool/args, tool_name:action, and method-shaped requests. This keeps new prompts clean without breaking agents that learned the previous dialect.
Move A0 connector remote execution/file tools into stable standard prompts, make remote targeting independent of the active chat context, and skill-gate beta computer-use remote so it no longer weighs down the always-on tool list.
Align text editor, scheduler, skills, office artifact, memory, notify, and browser prompts/tools around the canonical action contract. Add scheduler update/timezone handling, skills_tool read_file, text editor patch coverage, and fixes for memory_forget, behaviour_adjustment, and code execution progress warnings.
Reduce default prompt pressure by compacting browser and scheduler prompts into skill-backed manifests, shortening skill catalog descriptions, and pruning noisy framework knowledge. Remove obsolete connector prompt stubs and root tool-call knowledge examples.
Tests: conda run -n a0 pytest tests/test_a0_connector_prompt_gating.py tests/test_tool_action_contracts.py tests/test_task_scheduler_timezone.py tests/test_text_editor_context_patch.py tests/test_tool_request_normalization.py tests/test_office_document_store.py::test_odf_is_advertised_and_docx_remains_explicit_compatibility tests/test_office_document_store.py::test_document_artifact_accepts_method_alias_for_ods_create tests/test_skills_runtime.py tests/test_default_prompt_budget.py::test_a0_small_profile_removed_and_prompt_text_generic -q
Restyle Settings and standard modals around a streamlined left-rail layout, clearer section hierarchy, advanced settings disclosures, and stronger update states.
Add persistent update visibility with quieter once-daily update notifications, plus Remote Link and Space Agent actions in the canvas rail. Refresh the tunnel experience as a normal Remote Link modal with clearer copy, QR/mobile affordances, and safer state handling.
Add a builtin `a0-setup-cli` skill for guiding host-side A0 connector setup,
and restore the lightweight trigger-word based skill matching flow, which many users asked for.
- add builtin `skills/a0-setup-cli/` with installer-first host setup guidance,
container guardrails, fallback install paths, and example responses
- fix `helpers.skills_cli` so builtin skills under `/skills` are discoverable,
searchable, and validatable
- restore trigger-pattern scoring in runtime `search_skills()`
- re-enable `skills_tool:search` in the current tool flow
- add lightweight lexical relevant-skill recall for the current user message
without reintroducing memory/vector-db skill recall
- update skill prompts to steer the agent toward search/load when requests
match skill trigger phrases
- Refactor API key management: move from global env-only to per-model config with dotenv fallback
- Add API key placeholder masking and reveal functionality in WebUI
- Consolidate API key validation logic into `has_provider_api_key()` helper
- Improve update system: add branch filtering for tags, simplify backup naming
- Add branch detection to version info and default to current branch for updates
- Extract configure model settings link to constant
- Change @extensible decorator to generate extension paths from full module and qualname hierarchies instead of flattened names
- Update extension path format from `{module}_{qualname}_{start|end}` to `_functions/<module>/<qualname>/{start|end}`
- Move all extension files to new deep directory structure under `_functions/` to match new path format
- Replace PathSpec.from_lines(GitWildMatchPattern, ...) with PathSpec.from_lines
- Rename /state_sync namespace to /webui throughout codebase
- Remove get_event_types() from WebSocketHandler - handlers now process all events for their namespace
- Replace per-event handler registration with namespace-wide registration
- Add validate_event_type() class method for runtime event name validation
- Update UserMessage instantiation to use keyword arguments (message=, attachments=)
- Move send_data
- split monolithic _10_system_prompt.py into focused extensions: main (10), tools (11), mcp (12), skills (13), secrets (13), project (14)
- each extension exposes a build_prompt() function with call_extensions_async hook for plugin extensibility
- move tool prompt collection from VariablesPlugin to _11_tools_prompt using subagents.get_paths for proper directory coverage
- add {{include original}} directive to process_includes allowing prompt inheritance without copy-paste
- add agent.system.main.specifics.md for subagent-specific additions without overriding entire role
- remove redundant plugin prompt extensions (_15_text_editor, _20_code_execution) that duplicated tool collection
- add _09_text_editor_config to register per-file kwargs via agent.data instead of VariablesPlugin in prompts dir
- Move allow_chat_override from chat_model to top-level configuration
- Update all references to use new config path
- Set default to true in default_config.yaml, false in migration
- Fix datetime parsing to handle timezone-less ISO strings by appending 'Z'
- Remove inline style from tune icon in config.html
- Add `watchdog` dependency to requirements.txt
- Implement cache entry timestamps with `CacheEntry` dataclass for LRU-style tracking
- Add `trim_cache` function to remove stale entries based on age
- Update cache operations to track and update entry timestamps on access
- Add `_get_matching_areas` helper for wildcard pattern matching in cache clearing
- Register watchdogs for API handlers, extensions, and plugins to
- Extract presets into global presets.yaml; move editor to plugin main screen
- Add project-wide override sync and three-layer permission checks
- Fix embedding change detection, new-chat inheritance, and preset name leak bugs
- Clean up dead imports/params and fix description text
Redesign extension handling to support explicit async/sync execution. helpers/extension.py rewrites the extensible decorator, adds call_extensions_async / call_extensions_sync, and a helper to gather extension classes; caching flag adjusted. Updated call sites across the codebase (agent, APIs, plugins, tools, settings, extensions) to use extension.extensible and the new call_extensions_async/sync API, and converted several extension handlers from async to sync. Also small frontend tweaks (use globalThis.runtimeInfo) and minor import updates (csrf_protect in run_ui). This centralizes extension discovery/execution and avoids previously scattered asyncio.run usage.
