You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: AGENTS.md
+1Lines changed: 1 addition & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -264,6 +264,7 @@ The gate has a **second pass** for sources that emit `sleep_unspecified` only (n
264
264
|`REPORT_EVENING_WEEKEND`|`21`| Evening report hour on weekends |
265
265
|`REPORT_MORNING_CAP`|`morningHour+4` (floor 11) | Smart-retry deadline for morning report — past this hour the report force-sends with a stale-data banner regardless of `SleepSettled`. Server prefers an adaptive cap from `GetTypicalWakeTime(14) + 60min` when ≥7 days of per-segment sleep data are available; falls back to this env value otherwise. Override via env or `report_morning_cap` setting. **Floor:**`MorningCapTime` pushes the cap to `now + MinPromptWindow` (60min) whenever the computed cap is already in the past at call time — adaptive cap can land before `morning_hour` for users whose schedule puts the morning report well after typical wake, and without the floor the smart-retry loop enters past cap on first tick and silently skips `MorningActionPrompt`, disabling subjective check-in entirely. Pinned by `TestMorningCapTime_FloorsPastCapsToPromptWindow`|
266
266
|`REPORT_TZ`| system local | Timezone for report scheduling (e.g. `Europe/Belgrade`) |
267
+
|`TELEGRAM_RICH_MESSAGES`|`false`| Opt-in Telegram Bot API Rich Messages for morning/evening reports. DB setting `telegram_rich_messages` uses the same boolean semantics. Fallback stays on legacy `sendMessage` HTML when rich send fails. |
267
268
|`GEMINI_API_KEY`| — | Gemini API key; enables AI morning briefing |
268
269
|`GEMINI_MODEL`|`gemini-2.5-flash`| Gemini model; overridable in Admin UI |
269
270
|`GEMINI_MAX_TOKENS`|`5000`| Max output tokens for AI briefing; overridable in Admin UI |
Copy file name to clipboardExpand all lines: CLAUDE.md
+1Lines changed: 1 addition & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -264,6 +264,7 @@ The gate has a **second pass** for sources that emit `sleep_unspecified` only (n
264
264
|`REPORT_EVENING_WEEKEND`|`21`| Evening report hour on weekends |
265
265
|`REPORT_MORNING_CAP`|`morningHour+4` (floor 11) | Smart-retry deadline for morning report — past this hour the report force-sends with a stale-data banner regardless of `SleepSettled`. Server prefers an adaptive cap from `GetTypicalWakeTime(14) + 60min` when ≥7 days of per-segment sleep data are available; falls back to this env value otherwise. Override via env or `report_morning_cap` setting. **Floor:**`MorningCapTime` pushes the cap to `now + MinPromptWindow` (60min) whenever the computed cap is already in the past at call time — adaptive cap can land before `morning_hour` for users whose schedule puts the morning report well after typical wake, and without the floor the smart-retry loop enters past cap on first tick and silently skips `MorningActionPrompt`, disabling subjective check-in entirely. Pinned by `TestMorningCapTime_FloorsPastCapsToPromptWindow`|
266
266
|`REPORT_TZ`| system local | Timezone for report scheduling (e.g. `Europe/Belgrade`) |
267
+
|`TELEGRAM_RICH_MESSAGES`|`false`| Opt-in Telegram Bot API Rich Messages for morning/evening reports. DB setting `telegram_rich_messages` uses the same boolean semantics. Fallback stays on legacy `sendMessage` HTML when rich send fails. |
267
268
|`GEMINI_API_KEY`| — | Gemini API key; enables AI morning briefing |
268
269
|`GEMINI_MODEL`|`gemini-2.5-flash`| Gemini model; overridable in Admin UI |
269
270
|`GEMINI_MAX_TOKENS`|`5000`| Max output tokens for AI briefing; overridable in Admin UI |
0 commit comments