@herjarsa/omo-meta-governor 0.49.0 → 0.49.1

package/README.md

@@ -602,6 +602,9 @@ All configuration lives under the `meta_governor` key in
 |-------|------|---------|-------------|
 | `saveDecisions` | boolean | `true` | Whether to save decision records. |
 | `saveLessons` | boolean | `true` | Whether to save lessons. |
+| `autoRemember.enabled` | boolean | `true` | v0.49.1: kill-switch for the Phase 4 auto-remember prompt (warn/escalate/stop → agentmemory_memory_save). Set `false` to disable it entirely. |
+| `autoRemember.cooldownMs` | number | `300000` | v0.49.1: min ms between two auto-remember fires in the same session. |
+| `autoRemember.dedupe` | boolean | `true` | v0.49.1: identical content is never re-queued in the same session. |
 
 ### `modelOverride`
 

package/dist/config.d.ts

@@ -58,6 +58,12 @@ export interface MetaGovernorPluginConfig {
         maxLessonsPerSession?: number;
         saveDecisions?: boolean;
         saveLessons?: boolean;
+        /** v0.49.1: anti-loop guard for the Phase 4 auto-remember prompt. */
+        autoRemember?: {
+            enabled?: boolean;
+            cooldownMs?: number;
+            dedupe?: boolean;
+        };
     };
     /** Model override for MetaGovernor internal LLM usage. */
     modelOverride?: ModelOverrideConfig;