Till JS 3edf680ea0 feat(mana-ai): telemetry for reminder producers (mana_ai_reminders_emitted_total) ... Producers now return structured {producer, severity, text} objects instead of raw strings. buildReminderChannel collects them, increments mana_ai_reminders_emitted_total{producer, severity} per emission, and maps back to strings for the shared-ai loop input. Why structured: the Prometheus label "severity" lets dashboards split 75-99% token-budget warnings (severity=warn) from 100%+ escalations (severity=escalate) without NLP on the reminder text. Adding a new producer that emits only info-level state (e.g. stale-sync warning) falls out for free. Active producer labels today: - token-budget (warn, escalate) - retry-loop (warn) With this plus the scrape job (d087b4744), we can finally answer: "does the budget warning actually change LLM behaviour?" — correlate reminders_emitted_total{producer='token-budget'} with tick_duration_seconds or planner_rounds_histogram. 3 tests updated to assert the new {producer, severity, text} shape (16 reminder tests total, all green). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>		2026-04-23 15:10:27 +02:00
..
src	feat(mana-ai): telemetry for reminder producers (mana_ai_reminders_emitted_total)	2026-04-23 15:10:27 +02:00
CLAUDE.md	feat(agent-loop): activate retryLoopReminder via LoopState.recentCalls	2026-04-23 15:02:40 +02:00
Dockerfile	fix(infra): include shared-logger in mana-ai + mana-auth Dockerfile installers	2026-04-15 14:34:08 +02:00
package.json	feat(agent-loop): M1 — policy gate + reminder channel + parallel reads	2026-04-23 13:56:40 +02:00
tsconfig.json	feat(agent-loop): M1 — policy gate + reminder channel + parallel reads	2026-04-23 13:56:40 +02:00