Add flex-wrap, max-h-20, and overflow-y-auto to the groups container
in UserStatus.svelte to prevent horizontal overflow when users have
many groups. Groups now wrap to multiple rows with a scrollbar when
exceeding the max height.
Co-authored-by: QinQin <qinqin@geotab.com>
Add AUDIT_INCLUDED_PATHS env var for whitelist-based audit filtering.
When set, only matching paths are audited and AUDIT_EXCLUDED_PATHS is
ignored. Auth endpoints (signin/signout/signup) are always logged
regardless of filtering mode.
* Prefer model-provided web search result count over admin default
Update `search_web` to prioritize the model-provided `count` parameter before falling back to the admin-configured `WEB_SEARCH_RESULT_COUNT`, and finally defaulting to 5.
Changes:
- Set `count` default to `None` instead of `5`.
- Adjust fallback order to: model-provided `count` → admin-configured value → `5`.
- Update comment to reflect the new precedence logic.
This ensures explicit model requests for result count are respected while preserving sensible defaults.
* Enforce maximum web search result count from config
Update `search_web` to cap the model-provided `count` parameter at the admin-configured `WEB_SEARCH_RESULT_COUNT` to prevent excessive result requests.
Changes:
- Set default `count` parameter to `5`.
- Replace fallback logic with enforcement logic that limits `count` to the configured maximum.
- Update comment to reflect that the result count is now capped to prevent abuse.
This ensures web search requests cannot exceed the configured limit while maintaining a sensible default.
Some databases contain chat records where 'history' or 'messages' are
stored as lists instead of dicts. This causes an AttributeError
('list' object has no attribute 'items') during the
8452d01d26d7_add_chat_message_table migration.
Add isinstance checks to skip chat records with unexpected data shapes
gracefully, matching the existing pattern used for individual message
validation.
Rework shimmer animation timing and gradient for a smoother sweep.
Slower start with acceleration through the text using a custom
cubic-bezier curve. Softer highlight contrast in both light and
dark modes.
Add |middletruncate:n, |start:n, and |end:n pipe filters to the
{{MESSAGES}} template variable, enabling per-message character
truncation for task models (title, tags, follow-up, etc.).
Example: {{MESSAGES:END:2|middletruncate:500}}
This optimizes task model prompt size for conversations with very
long messages (e.g. pasted documents), reducing latency for local
models and API costs.
Closes#21499
