@flowingspring/dsh-voco 0.3.5 → 0.3.7

package/lib/plugins/voice-assistant.js

@@ -5720,6 +5720,31 @@ function apply(ctx, config = {}) {
 			ctx.logger.warn(error instanceof Error ? error : new Error(String(error)));
 		});
 	};
+	const submitTaskMemory = (task) => {
+		if (task.interactionMode !== "frontend-agent") return;
+		const memory = optionalWorkspaceMemory(ctx);
+		if (memory === void 0) return;
+		const request = task.requestText.trim();
+		const result = (task.completionDetail ?? task.lastAssistantMessage?.text ?? "").trim();
+		const prefix = `voice-task:${String(task.id)}`;
+		const messages = [...request === "" ? [] : [{
+			id: `${prefix}:request`,
+			role: "user",
+			text: request
+		}], ...result === "" ? [] : [{
+			id: `${prefix}:result`,
+			role: "assistant",
+			text: result
+		}]];
+		if (messages.length === 0) return;
+		memory.checkpoint({
+			sessionId: task.taskSessionId,
+			messages,
+			reason: "task-end"
+		}).catch((error) => {
+			ctx.logger.warn(error instanceof Error ? error : new Error(String(error)));
+		});
+	};
 	const markVoiceTurn = (binding) => {
 		if (binding.interactionMode !== "frontend-agent" || binding.voiceTurnMarked) return;
 		const session = requireSourceSession(binding);
@@ -6325,12 +6350,14 @@ function apply(ctx, config = {}) {
 		}
 		const deliverAssistantSpeech = !hasLlm || !ctx.voice.supportsSpeechText(binding.voiceSessionId);
 		enqueue(binding, () => {
+			const terminalTask = event.type === "turn/end" && event.data.reason.kind === "completed" && binding.active?.taskTurn === event.data.turn ? binding.active : void 0;
 			if (observeSessionEvent(binding, event, append, config, (error) => {
 				ctx.logger.warn(error instanceof Error ? error : new Error(String(error)));
 			}, (task, text) => {
 				if (deliverAssistantSpeech) ctx.voice.requestResponse(binding.voiceSessionId, { kind: "automatic" });
 				else rewriteAndSpeak(binding, task.id, task.requestText, text);
 			}, deliverAssistantSpeech)) {
+				if (terminalTask !== void 0) submitTaskMemory(terminalTask);
 				if ((config.taskSessionPolicy ?? "isolated") === "isolated") taskBindings.delete(session.id);
 			}
 		});

package/lib/plugins/voice-local.js

@@ -5,6 +5,8 @@ import { existsSync } from "node:fs";
 import { dirname, join, resolve } from "node:path";
 import { loadEnvFile } from "node:process";
 import { fileURLToPath } from "node:url";
+import { credentialRef } from "@deepseek-ai/dsh-credentials";
+import { installSettingsSection, settingsNamespace } from "@deepseek-ai/dsh-settings";
 import { mkdtemp, readFile, rm } from "node:fs/promises";
 import { tmpdir } from "node:os";
 import { EdgeTTS } from "node-edge-tts";
@@ -539,8 +541,10 @@ function errorMessage(error) {
 /** Silence-gated SiliconFlow cloud ASR with Edge TTS. */
 const name = "voice-local";
 const inject = ["voice"];
+const VOICE_LOCAL_SETTINGS_NAMESPACE = settingsNamespace("voice-local");
+const SILICONFLOW_API_KEY_REF = credentialRef("SILICONFLOW_API_KEY");
 const Config = z.object({
-	apiKey: z.string(),
+	apiKey: z.string().role("secret"),
 	endpoint: z.string().default("https://api.siliconflow.cn/v1/audio/transcriptions"),
 	model: z.string().default("XingChenAGI/XingChenASR-V3.2-Ultra"),
 	interactionMode: z.union(["speech-shell", "frontend-agent"]).default("speech-shell"),
@@ -575,26 +579,34 @@ function loadProjectEnv() {
 	}
 }
 function apply(ctx, config = {}) {
-	loadProjectEnv();
+	if (ctx.get("credentials") === void 0) loadProjectEnv();
+	let current = () => config;
+	installSettingsSection(ctx, VOICE_LOCAL_SETTINGS_NAMESPACE, Config, config, {
+		setSource: (source) => {
+			current = source;
+		},
+		onChange: () => {}
+	});
 	return ctx.voice.registerProvider({
 		id: "local",
 		available: () => true,
 		connect: async ({ voiceSessionId, emit }) => {
+			const active = current();
 			const session = new LocalSession(new NodeSpeechBackend({
-				apiKey: config.apiKey ?? process.env.SILICONFLOW_API_KEY ?? "",
-				endpoint: config.endpoint ?? "https://api.siliconflow.cn/v1/audio/transcriptions",
-				model: config.model ?? "XingChenAGI/XingChenASR-V3.2-Ultra",
-				requestTimeoutMs: config.requestTimeoutMs ?? 6e4,
-				inputSampleRate: config.inputSampleRate ?? 16e3,
-				outputSampleRate: config.outputSampleRate ?? 48e3,
-				ttsRate: config.ttsRate ?? "+20%",
-				silenceDurationMs: config.silenceDurationMs ?? 1500,
-				speechThreshold: config.speechThreshold ?? .015,
-				minSpeechDurationMs: config.minSpeechDurationMs ?? 250,
-				preRollMs: config.preRollMs ?? 400,
-				trailingSilenceMs: config.trailingSilenceMs ?? 200,
-				maxUtteranceMs: config.maxUtteranceMs ?? 6e4
-			}), emit, voiceSessionId, config.interactionMode ?? "speech-shell");
+				apiKey: await resolveSiliconFlowApiKey(ctx, active),
+				endpoint: active.endpoint ?? "https://api.siliconflow.cn/v1/audio/transcriptions",
+				model: active.model ?? "XingChenAGI/XingChenASR-V3.2-Ultra",
+				requestTimeoutMs: active.requestTimeoutMs ?? 6e4,
+				inputSampleRate: active.inputSampleRate ?? 16e3,
+				outputSampleRate: active.outputSampleRate ?? 48e3,
+				ttsRate: active.ttsRate ?? "+20%",
+				silenceDurationMs: active.silenceDurationMs ?? 1500,
+				speechThreshold: active.speechThreshold ?? .015,
+				minSpeechDurationMs: active.minSpeechDurationMs ?? 250,
+				preRollMs: active.preRollMs ?? 400,
+				trailingSilenceMs: active.trailingSilenceMs ?? 200,
+				maxUtteranceMs: active.maxUtteranceMs ?? 6e4
+			}), emit, voiceSessionId, active.interactionMode ?? "speech-shell");
 			try {
 				await session.start();
 				return session;
@@ -605,5 +617,12 @@ function apply(ctx, config = {}) {
 		}
 	});
 }
+/** Resolve the key per connection so a settings-page write reaches the next attempt. */
+async function resolveSiliconFlowApiKey(ctx, config) {
+	if (config.apiKey !== void 0 && config.apiKey.trim() !== "") return config.apiKey;
+	const credentials = ctx.get("credentials");
+	if (credentials !== void 0) return (await credentials.resolve(SILICONFLOW_API_KEY_REF))?.value ?? "";
+	return process.env.SILICONFLOW_API_KEY ?? "";
+}
 //#endregion
-export { Config, LocalSession, NodeSpeechBackend, SiliconFlowAsr, apply, inject, name, pcm16MonoWav };
+export { Config, LocalSession, NodeSpeechBackend, SILICONFLOW_API_KEY_REF, SiliconFlowAsr, VOICE_LOCAL_SETTINGS_NAMESPACE, apply, inject, name, pcm16MonoWav, resolveSiliconFlowApiKey };

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@flowingspring/dsh-voco",
   "description": "Persistent voice conversations for DSH with cloud speech recognition, Edge TTS, and background Agent delegation",
-  "version": "0.3.5",
+  "version": "0.3.7",
   "keywords": [
     "deepseek-harness",
     "dsh",
@@ -65,6 +65,8 @@
         "@deepseek-ai/dsh-client-connection",
         "@deepseek-ai/dsh-client-locale",
         "@deepseek-ai/dsh-client-runtime",
+        "@deepseek-ai/dsh-client-ui-settings",
+        "@deepseek-ai/dsh-client-ui-settings-plugins",
         "@deepseek-ai/dsh-client-ui-conversation",
         "@deepseek-ai/dsh-client-ui-layout",
         "@deepseek-ai/dsh-client-ui-sidebar"
@@ -86,15 +88,19 @@
     "@deepseek-ai/dsh-client-connection": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-client-locale": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-client-runtime": "^0.1.1-rc.1",
+    "@deepseek-ai/dsh-client-ui-settings": "^0.1.1-rc.2",
+    "@deepseek-ai/dsh-client-ui-settings-plugins": "^0.1.1-rc.2",
     "@deepseek-ai/dsh-client-ui-conversation": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-client-ui-layout": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-client-ui-sidebar": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-client-ui-slots": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-host-webserver": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-invariants": "^0.1.1-rc.1",
+    "@deepseek-ai/dsh-credentials": "^0.1.1-rc.2",
     "@deepseek-ai/dsh-llm": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-session": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-session-persistence": "^0.1.1-rc.1",
+    "@deepseek-ai/dsh-settings": "^0.1.1-rc.2",
     "@deepseek-ai/dsh-system-prompt": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-tools": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-workspace": "^0.1.1-rc.1",
@@ -109,15 +115,19 @@
     "@deepseek-ai/dsh-client-connection": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-client-locale": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-client-runtime": "^0.1.1-rc.1",
+    "@deepseek-ai/dsh-client-ui-settings": "^0.1.1-rc.2",
+    "@deepseek-ai/dsh-client-ui-settings-plugins": "^0.1.1-rc.2",
     "@deepseek-ai/dsh-client-ui-conversation": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-client-ui-layout": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-client-ui-sidebar": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-client-ui-slots": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-host-webserver": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-invariants": "^0.1.1-rc.1",
+    "@deepseek-ai/dsh-credentials": "^0.1.1-rc.2",
     "@deepseek-ai/dsh-llm": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-session": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-session-persistence": "^0.1.1-rc.1",
+    "@deepseek-ai/dsh-settings": "^0.1.1-rc.2",
     "@deepseek-ai/dsh-system-prompt": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-tools": "^0.1.1-rc.1",
     "@deepseek-ai/dsh-workspace": "^0.1.1-rc.1",