Deep-merge model preset slots with the active configuration so custom context windows, rate limits, and nested kwargs survive preset switches.
Treat legacy utility preset defaults as implicit values, allow omitted utility and embedding slots to inherit configured models, and document the partial-preset behavior.
- Move load_classes_from_file and load_classes_from_folder from extract_tools to new modules helper
- Update all imports across api, extension, files, and plugins to use helpers.modules
- Add namespace purging to refresh_plugin_modules for selective plugin reload on Python changes
- Implement embed trimming in history based on model config max_embeds and vision support
- Add pre_update hook documentation to plugin
- Remove browser_http_headers from model config (config, migration, UI, helpers)
- Add safe_call function to extract_tools.py to safely invoke functions with filtered args/kwargs
- Update call_plugin_hook to use safe_call for better parameter handling
- Add _apply_defaults_from_env to apply environment variable defaults to plugin configs
- Delete additional plugin asset folders when deleting plugins
- Remove
- Reset presets to system defaults when "Default" is clicked in Configure Models (resetGlobalPresets() + backend reset action)
- Expand preset fields to full config parity: ctx_length, ctx_history/ctx_input, vision, rate limits, kwargs, browser_http_headers
- Extract _normalizePresets() helper inside store to eliminate duplication
- Hide "Edit Presets" link when Per-Chat Override is disabled
- Add Settings + API Keys navigation buttons to Model Presets page
- Add dynamic import guard for plugin-settings-store.js in cross-modal navigation buttons
- Rename model switcher label from "Default" to "Default LLM"
- 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
