@clowder-ai/companion 0.1.0-alpha.12 → 0.1.0-alpha.13

package/README.md

@@ -10,10 +10,19 @@ visible and ends the call in one click. Screen sharing has its own explicit
 picker and termination badge. There is no microphone action on a double-click,
 page load, body tap or drag.
 
+When a compatible Host advertises receive-only audio, **只听** opens the same
+conversation with playback but never requests or toggles a microphone. Older
+Hosts omit that capability, so the control stays hidden and ordinary voice keeps
+the exact legacy connect request. Text remains available independently.
+
 Typing works with voice off. The Host uses the existing owner conversation,
 configured duty cat, ordinary message routing and idempotency. The small history
 view reads a bounded recent subset; **完整聊天** opens the canonical conversation.
 No second transcript database, credentials or selectable Host identity live here.
+When the Host supplies a valid per-message identity snapshot, each history row
+keeps its real author while separately showing the partner, Live carrier and
+deep cat saved with that message. Legacy rows omit this context instead of
+borrowing today's selected companion.
 
 ## Host integration
 

package/manifest.json

@@ -1,6 +1,6 @@
 {
   "pluginId": "official.companion",
-  "version": "0.1.0-alpha.12",
+  "version": "0.1.0-alpha.13",
   "contractVersion": "0.1.0",
   "name": "猫猫球",
   "icon": {
@@ -40,7 +40,7 @@
       "bridgeVersion": "1.3.0",
       "surface": {
         "entrypoint": "renderer/index.html",
-        "integrity": "sha256-x3ponGs/dhxPH5D1rT9JrLqJpFA6rxrgMjSImhksoHQ="
+        "integrity": "sha256-vGmfNZUU2zOlDNsQUqPjrvcwmO6p+Qdi3aPwlI8Wcyc="
       },
       "presentation": {
         "width": 300,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@clowder-ai/companion",
-  "version": "0.1.0-alpha.12",
+  "version": "0.1.0-alpha.13",
   "description": "The desktop voice body for your Clowder companion",
   "license": "MIT",
   "type": "module",
@@ -26,8 +26,8 @@
     "prepack": "pnpm build"
   },
   "devDependencies": {
-    "@clowder-ai/plugin-contract": "0.1.0-beta.20",
-    "@clowder-ai/plugin-sdk": "0.1.0-beta.16"
+    "@clowder-ai/plugin-contract": "0.1.0-beta.21",
+    "@clowder-ai/plugin-sdk": "0.1.0-beta.17"
   },
   "repository": {
     "type": "git",

package/plugin.yaml

@@ -1,6 +1,6 @@
 {
   "pluginId": "official.companion",
-  "version": "0.1.0-alpha.12",
+  "version": "0.1.0-alpha.13",
   "contractVersion": "0.1.0",
   "name": "猫猫球",
   "icon": {
@@ -40,7 +40,7 @@
       "bridgeVersion": "1.3.0",
       "surface": {
         "entrypoint": "renderer/index.html",
-        "integrity": "sha256-x3ponGs/dhxPH5D1rT9JrLqJpFA6rxrgMjSImhksoHQ="
+        "integrity": "sha256-vGmfNZUU2zOlDNsQUqPjrvcwmO6p+Qdi3aPwlI8Wcyc="
       },
       "presentation": {
         "width": 300,

package/renderer/client.mjs

@@ -19,7 +19,7 @@ export function createCompanionClient(bridge) {
     return {
         state: () => invoke({ kind: 'state' }, 'state'),
         prepare: () => invoke({ kind: 'prepare' }, 'state'),
-        connectAudio: () => invoke({ kind: 'audio.connect' }, 'ok'),
+        connectAudio: (mode) => invoke(mode === undefined ? { kind: 'audio.connect' } : { kind: 'audio.connect', mode }, 'ok'),
         closeAudio: () => invoke({ kind: 'audio.close' }, 'ok'),
         muteMicrophone: (muted) => invoke({ kind: 'audio.microphone', muted }, 'ok'),
         muteSpeaker: (muted) => invoke({ kind: 'audio.speaker', muted }, 'ok'),

package/renderer/companion-identity.mjs

@@ -1,12 +1,13 @@
 const named = value => typeof value === 'string' && value.trim().length > 0;
 const actor = value => value && named(value.catId) && named(value.displayName);
 const model = value => value === null || (named(value) && value.length <= 160);
+const snapshotActor = value => actor(value) && value.catId.length <= 160 && value.displayName.length <= 160;
 
 function validSnapshot(value) {
   return value?.v === 1 && value.name === '猫猫球'
-    && actor(value.partner) && named(value.partner.skin)
-    && actor(value.live) && value.live.transport === 'gpt_live_v3' && model(value.live.verifiedModel)
-    && actor(value.deep) && value.deep.catId === value.partner.catId && model(value.deep.verifiedModel);
+    && snapshotActor(value.partner) && named(value.partner.skin) && value.partner.skin.length <= 160
+    && snapshotActor(value.live) && value.live.transport === 'gpt_live_v3' && model(value.live.verifiedModel)
+    && snapshotActor(value.deep) && value.deep.catId === value.partner.catId && model(value.deep.verifiedModel);
 }
 
 function display(snapshot, partnerLabel) {

package/renderer/conversation.mjs

@@ -7,26 +7,41 @@ export class CompanionConversation {
     this.active = false;
     this.generation = 0;
     this.phase = 'idle';
+    this.audioMode = null;
+    this.retryMode = null;
     this.muted = false;
     this.silent = false;
     this.failed = false;
   }
   show(message) {
     this.message = message;
-    this.render({ phase: this.phase, identity: this.identity, muted: this.muted, silent: this.silent, failed: this.failed, message });
+    this.render({ phase: this.phase, identity: this.identity, audioMode: this.audioMode, retryMode: this.retryMode,
+      muted: this.muted, silent: this.silent, failed: this.failed, message });
   }
   current(generation) { return this.active && generation === this.generation; }
-  listening() { return this.muted ? '麦克风已静音' : '正在听，直接说话就好'; }
-  async begin() {
+  listening() {
+    if (this.audioMode === 'receive_only') return '只听模式 · 麦克风未启用';
+    return this.muted ? '麦克风已静音' : '正在听，直接说话就好';
+  }
+  retrying(prefix) {
+    return this.audioMode === 'receive_only'
+      ? `${prefix} · 麦克风未启用，点击只听重试`
+      : `${prefix} · 点击语音聊重试`;
+  }
+  async begin(mode = 'duplex') {
     if (this.active || this.stopping || this.changingDocuments) return;
+    if (mode === 'receive_only' && !this.identity?.audio?.supportedModes?.includes(mode)) return;
+    if (!['duplex', 'receive_only'].includes(mode)) return;
     this.active = true;
+    this.audioMode = mode;
+    this.retryMode = null;
     this.failed = false;
     this.phase = 'connecting';
     const generation = ++this.generation;
     // Must reach the isolated preload while the click still has user activation.
     const preparation = this.client.prepare();
     this.show('正在连接…');
-    this.deadline = setTimeout(() => { if (this.current(generation)) void this.end('连接超时 · 点击语音聊重试', true); }, 60_000);
+    this.deadline = setTimeout(() => { if (this.current(generation)) void this.end(this.retrying('连接超时'), true); }, 60_000);
     try {
       const peer = this.createPeer(event => {
         if (!this.current(generation)) return;
@@ -37,12 +52,14 @@ export class CompanionConversation {
         if (event.type === 'error') void this.end(explainError(event), true);
       });
       this.peer = peer;
-      peer.muteMic(this.muted); peer.muteSpeaker(this.silent);
+      if (mode === 'duplex') peer.muteMic(this.muted);
+      peer.muteSpeaker(this.silent);
       // Submit capture intent while this same click is active. The Host waits
       // for its matching preparation before creating the media document.
-      const [identity] = await Promise.all([preparation, peer.connect()]);
+      const [identity] = await Promise.all([preparation, peer.connect(mode === 'receive_only' ? mode : undefined)]);
       if (!this.current(generation)) { await peer.close(); return; }
       if (identity.phase !== 'ready') throw { code: 'unavailable' };
+      if (mode === 'receive_only' && !identity.audio?.supportedModes?.includes(mode)) throw { code: 'unavailable' };
       this.identity = identity;
       this.show(this.message);
     } catch (error) {
@@ -56,30 +73,42 @@ export class CompanionConversation {
     clearTimeout(this.deadline);
     const peer = this.peer;
     this.peer = undefined;
+    const releasedMode = this.audioMode;
+    this.audioMode = null;
+    this.retryMode = this.failed ? releasedMode : null;
     this.phase = 'idle';
     this.show(message);
     await Promise.allSettled([peer?.close(), this.stopScreen()]);
   }
-  async end(message = '语音已结束 · 麦克风已关闭', failed = false) {
+  async end(message, failed = false) {
     if (this.stopping) return this.stopping;
+    const receiveOnly = this.audioMode === 'receive_only';
+    message ??= this.audioMode === 'receive_only' ? '只听已结束 · 麦克风未启用' : '语音已结束 · 麦克风已关闭';
     this.failed = failed;
     const local = this.releaseLocal(message);
     const operation = Promise.all([local, this.client.stop()]);
     this.stopping = operation;
     try { await operation; }
-    catch { this.show('麦克风已关闭 · 连接收尾尚未确认'); }
+    catch { this.show(receiveOnly ? '麦克风未启用 · 连接收尾尚未确认' : '麦克风已关闭 · 连接收尾尚未确认'); }
     finally { if (this.stopping === operation) this.stopping = undefined; }
   }
   async hostStopped(reason) {
     // Stop echoes must not erase a more useful local failure or manual ending.
     if (!this.active) return;
-    const messages = {
+    const receiveOnly = this.audioMode === 'receive_only';
+    const messages = receiveOnly ? {
+      locked: '屏幕已锁定 · 只听已停止，解锁后可点击只听继续',
+      suspended: '电脑已休眠 · 只听已停止，可点击只听继续',
+      hidden: '猫猫已收起 · 只听已停止',
+    } : {
       locked: '屏幕已锁定 · 语音已停止，解锁后可点击语音聊继续',
       suspended: '电脑已休眠 · 语音已停止，可点击语音聊继续',
       hidden: '猫猫已收起 · 语音已停止',
     };
     this.failed = !Object.hasOwn(messages, reason);
-    await this.releaseLocal(messages[reason] ?? '语音连接已中断 · 麦克风已关闭，点击语音聊重试');
+    await this.releaseLocal(messages[reason] ?? (receiveOnly
+      ? '只听连接已中断 · 麦克风未启用，点击只听重试'
+      : '语音连接已中断 · 麦克风已关闭，点击语音聊重试'));
   }
   async refresh() {
     if (this.refreshing) return;
@@ -89,8 +118,9 @@ export class CompanionConversation {
       const identity = await this.client.state();
       if (generation !== this.generation) return;
       this.identity = identity;
+      if (this.active && identity.audio?.activeMode) this.audioMode = identity.audio.activeMode;
       if (this.active && ['closed', 'failed', 'idle'].includes(identity.phase)) {
-        await this.end('语音连接已中断 · 点击语音聊重试', true);
+        await this.end(this.retrying('语音连接已中断'), true);
       } else this.show(this.message ?? '点开始聊天，直接对我说话');
     } catch (error) {
       if (generation === this.generation) {
@@ -134,6 +164,10 @@ export class CompanionConversation {
       return false;
     } finally { this.sending = false; }
   }
-  muteMic() { this.muted = !this.muted; this.peer?.muteMic(this.muted); this.show(this.phase === 'talking' ? this.listening() : this.message); }
+  muteMic() {
+    if (this.audioMode === 'receive_only') return;
+    this.muted = !this.muted; this.peer?.muteMic(this.muted);
+    this.show(this.phase === 'talking' ? this.listening() : this.message);
+  }
   muteSpeaker() { this.silent = !this.silent; this.peer?.muteSpeaker(this.silent); this.show(this.message); }
 }

package/renderer/index.html

@@ -4,6 +4,7 @@
 <body>
   <svg class="icon-definitions" aria-hidden="true">
     <symbol id="i-mic" viewBox="0 0 24 24"><rect x="9" y="3" width="6" height="12" rx="3"/><path d="M5 11a7 7 0 0 0 14 0M12 18v3M8 21h8"/></symbol>
+    <symbol id="i-listen" viewBox="0 0 24 24"><path d="M4 13v-1a8 8 0 0 1 16 0v1M4 13h3v7H6a2 2 0 0 1-2-2v-5ZM20 13h-3v7h1a2 2 0 0 0 2-2v-5Z"/></symbol>
     <symbol id="i-keyboard" viewBox="0 0 24 24"><rect x="2.5" y="5" width="19" height="14" rx="3"/><path d="M6 9h.01M10 9h.01M14 9h.01M18 9h.01M6 12h.01M10 12h.01M14 12h.01M18 12h.01M7 15h10"/></symbol>
     <symbol id="i-history" viewBox="0 0 24 24"><path d="M3 11a9 9 0 1 1 2.6 7.4M3 4v7h7M12 7v5l3 2"/></symbol>
     <symbol id="i-screen" viewBox="0 0 24 24"><rect x="3" y="4" width="18" height="13" rx="2.5"/><path d="M8 21h8M12 17v4"/></symbol>
@@ -36,7 +37,7 @@
         <video id="living-video-transition" muted playsinline preload="metadata" aria-hidden="true" hidden></video>
       </button>
       <div id="indicators">
-        <button id="call-badge" hidden aria-label="结束语音" title="结束语音"><svg viewBox="0 0 24 24" aria-hidden="true"><rect x="9" y="3" width="6" height="12" rx="3"/><path d="M5 11a7 7 0 0 0 14 0M12 18v4M8 22h8"/></svg></button>
+        <button id="call-badge" hidden aria-label="结束语音" title="结束语音"><svg viewBox="0 0 24 24" aria-hidden="true"><use id="call-badge-icon" href="#i-mic"/></svg></button>
         <button id="share-badge" hidden aria-label="停止共享" title="停止共享"><svg viewBox="0 0 24 24" aria-hidden="true"><rect x="3" y="4" width="18" height="13" rx="2"/><path d="M8 21h8M12 17v4"/></svg></button>
         <button id="pending-badge" hidden aria-label="查看待决事项" title="查看待决事项"><span id="pending-count">?</span></button>
       </div>
@@ -48,6 +49,7 @@
       <p id="status" role="status" hidden></p>
       <div class="action-row">
         <button id="begin" class="primary" data-action="begin"><svg class="icon" aria-hidden="true"><use href="#i-mic"/></svg><span class="label">语音聊</span></button>
+        <button id="listen" data-action="listen" hidden><svg class="icon" aria-hidden="true"><use href="#i-listen"/></svg><span class="label">只听</span></button>
         <button data-action="write"><svg class="icon" aria-hidden="true"><use href="#i-keyboard"/></svg><span>打字</span></button>
         <button id="mic" data-action="mute" hidden><svg class="icon" aria-hidden="true"><use href="#i-mic"/></svg><span class="label">静音</span></button>
       </div>

package/renderer/peer.mjs

@@ -10,7 +10,7 @@ export class VoicePeer {
       }
     });
   }
-  async connect() { if (!this.closed) await this.client.connectAudio(); }
+  async connect(mode) { if (!this.closed) await this.client.connectAudio(mode); }
   muteMic(muted) { void this.client.muteMicrophone(muted).catch(() => this.failed()); }
   muteSpeaker(muted) { void this.client.muteSpeaker(muted).catch(() => this.failed()); }
   failed() { if (!this.closed) this.onEvent({ type: 'error', code: 'unavailable' }); }

package/renderer/style.css

@@ -50,6 +50,7 @@ button:disabled { opacity: .45; cursor: default; }
 #indicators button:hover { background: #ab4842; }
 #indicators button[data-state="connecting"] { background: #a77b36; animation: pulse 1.8s infinite; }
 #indicators button[data-state="muted"] { background: #5f6670; }
+#indicators button[data-state="receive-only"] { background: #476dac; }
 #indicators svg { width: 16px; height: 16px; fill: none; stroke: currentColor; stroke-width: 1.8; stroke-linecap: round; }
 #share-badge { background: #476dac !important; }
 #pending-badge { background: #9b6841 !important; font-size: 11px; font-weight: 700; }
@@ -100,6 +101,7 @@ button:disabled { opacity: .45; cursor: default; }
 #pet[data-skin="yarn-ball"] { background-image: url("./skins/yarn-ball.png"); animation: none; background-size: contain; background-repeat: no-repeat; background-position: center; }
 #transcript { height: 230px; font-size: 13px; }
 #transcript p[data-author]::before { content: attr(data-author); display: block; color: var(--cafe-text-muted); font-size: 10px; margin-bottom: 2px; }
+#transcript p[data-companion]::after { content: attr(data-companion) " · " attr(data-companion-live) " · " attr(data-companion-deep); display: block; color: var(--cafe-text-muted); font-size: 9px; line-height: 1.35; margin-top: 4px; }
 #transcript p[data-live="true"] { border-left: 2px solid var(--cafe-accent); }
 #history { padding: 0; font-size: 10px; color: var(--cafe-text-secondary); }
 #decisions { width: 320px; min-height: 180px; padding: 8px 12px 12px; display: flex; flex-direction: column; overflow: hidden; }

package/renderer/surface.mjs

@@ -54,6 +54,7 @@ if (!window.clowderCompanion) {
     changed: panel => { if (panel === 'chat') void readHistory(); if (panel === 'decisions') void readDecisions(); },
     action(kind) {
       if (kind === 'begin') { transcript.reset(); bubble.reset(); void conversation.begin(); void controls.show('none'); }
+      if (kind === 'listen') { transcript.reset(); bubble.reset(); void conversation.begin('receive_only'); void controls.show('none'); }
       if (kind === 'stop') void conversation.end();
       if (kind === 'mute') conversation.muteMic();
       if (kind === 'speaker') conversation.muteSpeaker();
@@ -100,24 +101,35 @@ if (!window.clowderCompanion) {
       if (previousPhase !== 'idle' && value.phase === 'idle') { transcript.reset(); bubble.reset(); void readHistory(); }
       previousPhase = value.phase;
       const active = value.phase !== 'idle';
+      const receiveOnly = value.audioMode === 'receive_only';
+      const canReceiveOnly = value.identity?.audio?.supportedModes?.includes('receive_only') === true;
+      const retryReceiveOnly = value.failed && value.retryMode === 'receive_only' && canReceiveOnly;
       void controls.setAmbient(showAmbient(active));
       $('begin').hidden = active; $('begin').disabled = !value.identity;
-      label('begin', value.failed ? '重试语音' : '语音聊');
-      $('mic').hidden = value.phase !== 'talking'; $('speaker').hidden = !active;
+      $('listen').hidden = active || !canReceiveOnly; $('listen').disabled = !value.identity;
+      label('begin', value.failed && !retryReceiveOnly ? '重试语音' : '语音聊');
+      label('listen', retryReceiveOnly ? '重试只听' : '只听');
+      $('begin').className = retryReceiveOnly ? '' : 'primary';
+      $('listen').className = retryReceiveOnly ? 'primary' : '';
+      $('mic').hidden = value.phase !== 'talking' || receiveOnly; $('speaker').hidden = !active;
       $('active-actions').hidden = !active;
       $('share').disabled = value.phase !== 'talking';
       $('compose').querySelector('button').disabled = !value.identity || value.phase === 'connecting';
       status(value.message ?? '');
-      $('status').hidden = !value.message || /^(点|正在听|语音已结束|已发送)/.test(value.message);
+      $('status').hidden = !value.message || /^(点|正在听|只听模式|只听已结束|语音已结束|已发送)/.test(value.message);
       $('call-badge').hidden = !active;
-      $('call-badge').dataset.state = value.muted ? 'muted' : value.phase;
-      const callLabel = value.phase === 'connecting' ? '正在连接，点此取消' : value.muted ? '麦克风已静音，点此结束' : '语音进行中，点此结束';
+      $('call-badge').dataset.state = receiveOnly ? 'receive-only' : value.muted ? 'muted' : value.phase;
+      $('call-badge-icon').setAttribute('href', receiveOnly ? '#i-listen' : '#i-mic');
+      const callLabel = value.phase === 'connecting' ? (receiveOnly ? '正在连接只听，麦克风未启用，点此取消' : '正在连接，点此取消')
+        : receiveOnly ? '只听进行中，麦克风未启用，点此结束'
+          : value.muted ? '麦克风已静音，点此结束' : '语音进行中，点此结束';
       $('call-badge').title = callLabel; $('call-badge').setAttribute('aria-label', callLabel);
       label('mic', value.muted ? '取消静音' : '静音'); $('mic').setAttribute('aria-pressed', String(value.muted));
       label('speaker', value.silent ? '开启播音' : '关闭播音');
       const identity = value.identity;
       const companionIdentity = currentCompanionIdentity(identity);
-      motion.setContext({ skin: identity?.skin, phase: value.phase, nativeActivity: identity?.nativeActivity ?? 'none', muted: value.muted });
+      motion.setContext({ skin: identity?.skin, phase: value.phase,
+        nativeActivity: identity?.nativeActivity ?? 'none', muted: value.muted || receiveOnly });
       motion.syncNativeSnapshot(normalizeNativeWork(identity?.nativeWork));
       if (justConnected) motion.signal('connected');
       if (showFailure) motion.signal('failed');

package/renderer/transcript-view.mjs

@@ -1,3 +1,5 @@
+import { historicalCompanionIdentity } from './companion-identity.mjs';
+
 export class TranscriptView {
   constructor(container) {
     this.container = container;
@@ -57,7 +59,23 @@ export class TranscriptView {
       const row = previous.get(message.id) ?? this.container.ownerDocument.createElement('p');
       row.className = message.role; row.dataset.messageId = message.id;
       if (row.textContent !== message.text) row.textContent = message.text;
-      row.title = message.name; row.dataset.author = message.name;
+      const identity = historicalCompanionIdentity(message.companionIdentity, message.name);
+      row.dataset.author = identity.authorLabel;
+      if (identity.companionKnown) {
+        row.dataset.companion = identity.partnerLabel;
+        row.dataset.companionLive = identity.liveLabel;
+        row.dataset.companionDeep = identity.deepLabel;
+        row.dataset.companionAvatar = identity.avatarCatId;
+        row.dataset.companionSkin = identity.skin;
+        row.title = [identity.authorLabel, identity.partnerLabel, identity.liveLabel, identity.deepLabel].join('\n');
+        row.setAttribute('aria-label', [identity.authorLabel, message.text,
+          identity.partnerLabel, identity.liveLabel, identity.deepLabel].join('。'));
+      } else {
+        for (const key of ['companion', 'companionLive', 'companionDeep', 'companionAvatar', 'companionSkin'])
+          delete row.dataset[key];
+        row.title = identity.authorLabel;
+        row.removeAttribute?.('aria-label');
+      }
       this.historyRows.set(message.id, row);
       return row;
     });