feat: improve AI chat - system prompt, tool output, context window display (#17769)

⚠️ **AI-generated PR — not ready for review** ⚠️

cc @FelixMalfait

---

## Changes

### System prompt improvements
- Explicit skill-before-tools workflow to prevent the model from calling
tools without loading the matching skill first
- Data efficiency guidance (default small limits, use filters)
- Pluralized `load_skill` → `load_skills` for consistency with
`load_tools`

### Token usage reduction
- Output serialization layer: strips null/undefined/empty values from
tool results
- Lowered default `find_*` limit from 100 → 10, max from 1000 → 100

### System object tool generation
- System objects (calendar events, messages, etc.) now generate AI tools
- Only workflow-related and favorite-related objects are excluded

### Context window display fix
- **Bug**: UI compared cumulative tokens (sum of all turns) against
single-request context window → showed 100% after a few turns
- **Fix**: Track `conversationSize` (last step's `inputTokens`) which
represents the actual conversation history size sent to the model
- New `conversationSize` column on thread entity with migration

### Workspace AI instructions
- Support for custom workspace-level AI instructions

---------

Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
This commit is contained in:
Félix Malfait
2026-02-09 14:26:02 +01:00
committed by GitHub
parent 6c7c389785
commit 3216b634a3
105 changed files with 3245 additions and 1714 deletions
@@ -119,8 +119,10 @@ export const useAgentChat = (uiMessages: ExtendedUIMessage[]) => {
type UsageMetadata = {
inputTokens: number;
outputTokens: number;
cachedInputTokens: number;
inputCredits: number;
outputCredits: number;
conversationSize: number;
};
type ModelMetadata = {
contextWindowTokens: number;
@@ -133,11 +135,17 @@ export const useAgentChat = (uiMessages: ExtendedUIMessage[]) => {
if (isDefined(usage) && isDefined(model)) {
setAgentChatUsage((prev) => ({
lastMessage: {
inputTokens: usage.inputTokens,
outputTokens: usage.outputTokens,
cachedInputTokens: usage.cachedInputTokens,
inputCredits: usage.inputCredits,
outputCredits: usage.outputCredits,
},
conversationSize: usage.conversationSize,
contextWindowTokens: model.contextWindowTokens,
inputTokens: (prev?.inputTokens ?? 0) + usage.inputTokens,
outputTokens: (prev?.outputTokens ?? 0) + usage.outputTokens,
totalTokens:
(prev?.totalTokens ?? 0) + usage.inputTokens + usage.outputTokens,
contextWindowTokens: model.contextWindowTokens,
inputCredits: (prev?.inputCredits ?? 0) + usage.inputCredits,
outputCredits: (prev?.outputCredits ?? 0) + usage.outputCredits,
}));
@@ -23,16 +23,17 @@ const setUsageFromThread = (
thread: AgentChatThread,
setAgentChatUsage: SetterOrUpdater<AgentChatUsageState | null>,
) => {
const totalTokens = thread.totalInputTokens + thread.totalOutputTokens;
const hasUsageData = totalTokens > 0 && isDefined(thread.contextWindowTokens);
const hasUsageData =
(thread.conversationSize ?? 0) > 0 && isDefined(thread.contextWindowTokens);
setAgentChatUsage(
hasUsageData
? {
lastMessage: null,
conversationSize: thread.conversationSize ?? 0,
contextWindowTokens: thread.contextWindowTokens ?? 0,
inputTokens: thread.totalInputTokens,
outputTokens: thread.totalOutputTokens,
totalTokens,
contextWindowTokens: thread.contextWindowTokens ?? 0,
inputCredits: thread.totalInputCredits,
outputCredits: thread.totalOutputCredits,
}