@minicode-dev/chat 0.1.3 → 0.1.4

package/CHANGELOG.md

@@ -14,6 +14,9 @@ All notable changes to `@minicode-dev/chat` (owner: Minicode, project: LIRA).
 
 ### Changed
 
+- Incoming effects autoplay by default in visible conversations, with a 5-second cooldown and no window-focus requirement.
+- New incoming effects can replace an active animation; confetti now fills the panel with 220 responsive particles.
+
 - Effects respect mute, reduced motion, cooldowns and manual replay; old messages and duplicates do not autoplay.
 - Standalone development and release documentation. Host applications must update their served sound assets.
 

package/README.md

@@ -222,16 +222,16 @@ or use the existing `setChatAssetsBaseUrl` sound asset configuration. Copying th
 Open the **Efectos y zumbidos** button in the composer and click an effect to send it
 and play its animation. Each compact row shows sending/sent feedback; sound and automatic
 playback are controlled by two switches. Received
-effects appear as cards with manual replay; automatic playback is opt-in.
+effects appear as cards with manual replay; new received effects play automatically by default in the visible conversation.
 
 ```jsx
 const config = {
   effects: {
     enabled: true,
-    autoPlay: false,
+    autoPlay: true,
     sound: true,
     volume: 0.35,
-    cooldownMs: 30000,
+    cooldownMs: 5000,
     maxAgeMs: 10000,
     allowGroups: false,
     onPreferencesChange: (patch) => console.log('Save user preferences', patch),
@@ -247,13 +247,13 @@ alongside the versioned payload. Set `effects: false` to disable the feature.
 Audio requires browser permission to play after a user gesture and respects the chat's
 mute control. Effects stop when closed or when leaving the conversation; motion reduction
 uses a static alternative. History, duplicates and reconnection baselines do not trigger
-automatic playback. Automatic effects require a visible, focused conversation. Same-origin
+automatic playback. Automatic effects require a visible conversation in a visible tab. Same-origin
 tab deduplication is best effort. Receiving clients use a cooldown and ignore stale events.
 The server should enforce its own authorization and rate limits; browser controls alone
 cannot prevent abusive clients.
 
 In the demo, **Nuevas funciones** lets you simulate effects from Ana and configure playback.
-The demo permits effects in groups and shortens the cooldown to 1.5 seconds. None of these
+The demo permits effects in groups and uses the same 5-second cooldown. None of these
 features requires a paid audio API. Sharing full songs and managing a production emoji
 upload service are outside this implementation.
 
@@ -281,3 +281,14 @@ proposal; the nudge and confetti WAVs are the original sources of the approved r
 them separately. Generated files live in `demo/audio-samples/`, are ignored by Git and
 excluded from npm. Source generators are in `demo/audio-lab/`; see
 [AUDIO_EFFECTS.md](AUDIO_EFFECTS.md) in the source repository for the arrangement and limitations.
+
+### Fun chat defaults
+
+New incoming effects automatically animate in the visible conversation by default;
+manual replay is still available. The default cooldown is 5 seconds. Window focus is
+not required, but hidden tabs and old messages do not replay effects. A new permitted
+effect can replace the animation currently playing. Confetti fills the panel with
+220 colored particles and scales to its height. Reduced-motion preferences remain respected.
+Browsers can still block sound before the recipient interacts with the site; the visual
+effect does not depend on successful audio playback. Both participants must use an updated
+package, and explicitly saved `autoPlay: false` settings still disable automatic playback.