From 08812e541cb67c618a8392240c3b8a1dd3ac2991 Mon Sep 17 00:00:00 2001 From: iven Date: Thu, 23 Apr 2026 09:20:35 +0800 Subject: [PATCH] =?UTF-8?q?fix(identity):=20=E6=8E=A5=E9=80=9A=E8=BA=AB?= =?UTF-8?q?=E4=BB=BD=E4=BF=A1=E5=8F=B7=E6=8F=90=E5=8F=96=E4=B8=8E=E6=8C=81?= =?UTF-8?q?=E4=B9=85=E5=8C=96=20=E2=80=94=20=E5=AF=B9=E8=AF=9D=E4=B8=AD?= =?UTF-8?q?=E8=B5=B7=E5=90=8D=E8=B7=A8=E4=BC=9A=E8=AF=9D=E8=AE=B0=E5=BF=86?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 根因: 记忆提取管道(COMBINED_EXTRACTION_PROMPT)提取5种画像信号 但无身份信号(agent_name/user_name)，不存在从对话到AgentConfig.name 或IdentityFiles的写回路径。 修复内容: - ProfileSignals 增加 agent_name/user_name 字段 - COMBINED_EXTRACTION_PROMPT 增加身份提取指令 - parse_profile_signals 解析新字段 + 回退推断 - GrowthIntegration 存储身份信号到 VikingStorage - post_conversation_hook 写回 soul.md + emit Tauri 事件 - streamStore 规则化检测 agent 名字并更新 AgentConfig.name - cold-start-mapper 新增 detectAgentNameSuggestion 链路: 对话→提取→VikingStorage→hook写回soul.md→事件→前端刷新 --- crates/zclaw-growth/src/extractor.rs | 79 ++++++- crates/zclaw-growth/src/types.rs | 28 +++ crates/zclaw-runtime/src/growth.rs | 33 +++ desktop/src-tauri/src/intelligence_hooks.rs | 70 ++++++ desktop/src-tauri/src/kernel_commands/chat.rs | 5 +- desktop/src/lib/cold-start-mapper.ts | 26 +++ desktop/src/store/chat/streamStore.ts | 212 ++++++++++++++++-- 7 files changed, 431 insertions(+), 22 deletions(-) diff --git a/crates/zclaw-growth/src/extractor.rs b/crates/zclaw-growth/src/extractor.rs index 470a9ee..1fa28bf 100644 --- a/crates/zclaw-growth/src/extractor.rs +++ b/crates/zclaw-growth/src/extractor.rs @@ -253,6 +253,18 @@ impl MemoryExtractor { Ok(stored) } + /// Store a single pre-built MemoryEntry to VikingStorage + pub async fn store_memory_entry(&self, entry: &crate::types::MemoryEntry) -> Result<()> { + let viking = match &self.viking { + Some(v) => v, + None => { + tracing::warn!("[MemoryExtractor] No VikingAdapter configured"); + return Err(zclaw_types::ZclawError::Internal("No VikingAdapter".to_string())); + } + }; + viking.store(entry).await + } + /// 统一提取：单次 LLM 调用同时产出 memories + experiences + profile_signals /// /// 优先使用 `extract_with_prompt()` 进行单次调用；若 driver 不支持则 @@ -481,6 +493,16 @@ fn parse_profile_signals(obj: &serde_json::Value) -> crate::types::ProfileSignal .and_then(|s| s.get("communication_style")) .and_then(|v| v.as_str()) .map(String::from), + agent_name: signals + .and_then(|s| s.get("agent_name")) + .and_then(|v| v.as_str()) + .filter(|s| !s.is_empty()) + .map(String::from), + user_name: signals + .and_then(|s| s.get("user_name")) + .and_then(|v| v.as_str()) + .filter(|s| !s.is_empty()) + .map(String::from), } } @@ -525,6 +547,22 @@ fn infer_profile_signals_from_memories( signals.communication_style = Some(m.content.clone()); } } + // 身份信号回退: 从 preference 记忆中检测命名/称呼关键词 + let lower = m.content.to_lowercase(); + if lower.contains("叫你") || lower.contains("助手名字") || lower.contains("称呼") { + if signals.agent_name.is_none() { + // 尝试提取引号内的名字 + signals.agent_name = extract_quoted_name(&m.content) + .or_else(|| extract_name_after_pattern(&lower, &m.content, "叫你")); + } + } + if lower.contains("我叫") || lower.contains("我的名字") || lower.contains("用户名") { + if signals.user_name.is_none() { + signals.user_name = extract_name_after_pattern(&lower, &m.content, "我叫") + .or_else(|| extract_name_after_pattern(&lower, &m.content, "我的名字是")) + .or_else(|| extract_name_after_pattern(&lower, &m.content, "我叫")); + } + } } crate::types::MemoryType::Knowledge => { if signals.recent_topic.is_none() && !m.keywords.is_empty() { @@ -547,6 +585,38 @@ fn infer_profile_signals_from_memories( signals } +/// 从引号中提取名字（如"以后叫你'小马'"→"小马"） +fn extract_quoted_name(text: &str) -> Option { + for delim in ['"', '\'', '「', '」', '『', '』'] { + let mut parts = text.split(delim); + parts.next(); // skip before first delimiter + if let Some(name) = parts.next() { + let trimmed = name.trim(); + if !trimmed.is_empty() && trimmed.chars().count() <= 20 { + return Some(trimmed.to_string()); + } + } + } + None +} + +/// 从指定模式后提取名字（如"叫你小马"→"小马"） +fn extract_name_after_pattern(lower: &str, original: &str, pattern: &str) -> Option { + if let Some(pos) = lower.find(pattern) { + let after = &original[pos + pattern.len()..]; + // 取第一个词（中文或英文，最多10个字符） + let name: String = after + .chars() + .take_while(|c| !c.is_whitespace() && !matches!(c, '，'| '。' | '！' | '？' | ',' | '.' | '!' | '?')) + .take(10) + .collect(); + if !name.is_empty() { + return Some(name); + } + } + None +} + /// Default extraction prompts for LLM pub mod prompts { use crate::types::MemoryType; @@ -594,7 +664,9 @@ pub mod prompts { "recent_topic": "最近讨论的主要话题(可选)", "pain_point": "用户当前痛点(可选)", "preferred_tool": "用户偏好的工具/技能(可选)", - "communication_style": "沟通风格: concise|detailed|formal|casual(可选)" + "communication_style": "沟通风格: concise|detailed|formal|casual(可选)", + "agent_name": "用户给助手起的名称(可选，仅在用户明确命名时填写，如'以后叫你小马')", + "user_name": "用户提到的自己的名字(可选，仅在用户明确自我介绍时填写，如'我叫张三')" } } ``` @@ -604,8 +676,9 @@ pub mod prompts { 1. **memories**: 提取用户偏好(沟通风格/格式/语言)、知识(事实/领域知识/经验教训)、使用经验(技能/工具使用模式和结果) 2. **experiences**: 仅提取明确的"问题→解决"模式，要求有清晰的痛点和步骤，confidence >= 0.6 3. **profile_signals**: 从对话中推断用户画像信息，只在有明确信号时填写，留空则不填 -4. 每个字段都要有实际内容，不确定的宁可省略 -5. 只返回 JSON，不要附加其他文本 +4. **identity**: 检测用户是否给助手命名(如"你叫X"/"以后叫你X"/"你的名字是X")或自我介绍(如"我叫X"/"我的名字是X")，填入 agent_name 或 user_name 字段 +5. 每个字段都要有实际内容，不确定的宁可省略 +6. 只返回 JSON，不要附加其他文本 对话内容： "#; diff --git a/crates/zclaw-growth/src/types.rs b/crates/zclaw-growth/src/types.rs index e291a1a..c55f4ae 100644 --- a/crates/zclaw-growth/src/types.rs +++ b/crates/zclaw-growth/src/types.rs @@ -432,6 +432,10 @@ pub struct ProfileSignals { pub pain_point: Option, pub preferred_tool: Option, pub communication_style: Option, + /// 用户给助手起的名称（如"以后叫你小马"） + pub agent_name: Option, + /// 用户提到的自己的名字（如"我叫张三"） + pub user_name: Option, } impl ProfileSignals { @@ -442,6 +446,8 @@ impl ProfileSignals { || self.pain_point.is_some() || self.preferred_tool.is_some() || self.communication_style.is_some() + || self.agent_name.is_some() + || self.user_name.is_some() } /// 有效信号数量 @@ -452,8 +458,15 @@ impl ProfileSignals { if self.pain_point.is_some() { count += 1; } if self.preferred_tool.is_some() { count += 1; } if self.communication_style.is_some() { count += 1; } + if self.agent_name.is_some() { count += 1; } + if self.user_name.is_some() { count += 1; } count } + + /// 是否包含身份信号（agent_name 或 user_name） + pub fn has_identity_signal(&self) -> bool { + self.agent_name.is_some() || self.user_name.is_some() + } } /// 进化事件 @@ -674,8 +687,23 @@ mod tests { pain_point: None, preferred_tool: Some("researcher".to_string()), communication_style: Some("concise".to_string()), + agent_name: None, + user_name: None, }; assert_eq!(signals.industry.as_deref(), Some("healthcare")); assert!(signals.pain_point.is_none()); + assert!(!signals.has_identity_signal()); + } + + #[test] + fn test_profile_signals_identity() { + let signals = ProfileSignals { + agent_name: Some("小马".to_string()), + user_name: Some("张三".to_string()), + ..Default::default() + }; + assert!(signals.has_identity_signal()); + assert_eq!(signals.signal_count(), 2); + assert_eq!(signals.agent_name.as_deref(), Some("小马")); } } diff --git a/crates/zclaw-runtime/src/growth.rs b/crates/zclaw-runtime/src/growth.rs index ee301f3..1357014 100644 --- a/crates/zclaw-runtime/src/growth.rs +++ b/crates/zclaw-runtime/src/growth.rs @@ -440,6 +440,39 @@ impl GrowthIntegration { } } + // Store identity signals as special memories for cross-session persistence + if combined.profile_signals.has_identity_signal() { + let agent_id_str = agent_id.to_string(); + if let Some(ref agent_name) = combined.profile_signals.agent_name { + let entry = zclaw_growth::types::MemoryEntry::new( + &agent_id_str, + zclaw_growth::types::MemoryType::Preference, + "identity", + format!("助手的名字是{}", agent_name), + ).with_importance(8) + .with_keywords(vec!["名字".to_string(), "称呼".to_string(), "identity".to_string(), agent_name.clone()]); + if let Err(e) = self.extractor.store_memory_entry(&entry).await { + tracing::warn!("[GrowthIntegration] Failed to store agent_name signal: {}", e); + } else { + tracing::info!("[GrowthIntegration] Stored agent_name '{}' for {}", agent_name, agent_id_str); + } + } + if let Some(ref user_name) = combined.profile_signals.user_name { + let entry = zclaw_growth::types::MemoryEntry::new( + &agent_id_str, + zclaw_growth::types::MemoryType::Preference, + "identity", + format!("用户的名字是{}", user_name), + ).with_importance(8) + .with_keywords(vec!["名字".to_string(), "用户名".to_string(), "identity".to_string(), user_name.clone()]); + if let Err(e) = self.extractor.store_memory_entry(&entry).await { + tracing::warn!("[GrowthIntegration] Failed to store user_name signal: {}", e); + } else { + tracing::info!("[GrowthIntegration] Stored user_name '{}' for {}", user_name, agent_id_str); + } + } + } + // Convert extracted memories to structured facts let facts: Vec = combined .memories diff --git a/desktop/src-tauri/src/intelligence_hooks.rs b/desktop/src-tauri/src/intelligence_hooks.rs index 3bca918..7b6b057 100644 --- a/desktop/src-tauri/src/intelligence_hooks.rs +++ b/desktop/src-tauri/src/intelligence_hooks.rs @@ -8,6 +8,8 @@ use tracing::{debug, warn}; use std::sync::Arc; +use tauri::Emitter; +use zclaw_growth::VikingStorage; use crate::intelligence::identity::IdentityManagerState; use crate::intelligence::heartbeat::HeartbeatEngineState; @@ -56,12 +58,15 @@ pub async fn pre_conversation_hook( /// /// 1. Record interaction for heartbeat engine /// 2. Record conversation for reflection engine, trigger reflection if needed +/// 3. Detect identity signals and write back to identity files pub async fn post_conversation_hook( agent_id: &str, _user_message: &str, _heartbeat_state: &HeartbeatEngineState, reflection_state: &ReflectionEngineState, llm_driver: Option>, + identity_state: &IdentityManagerState, + app: &tauri::AppHandle, ) { // Step 1: Record interaction for heartbeat crate::intelligence::heartbeat::record_interaction(agent_id); @@ -200,6 +205,71 @@ pub async fn post_conversation_hook( reflection_result.improvements.len() ); } + + // Step 3: Detect identity signals from recent memory extraction and write back + if let Ok(storage) = crate::viking_commands::get_storage().await { + let identity_prefix = format!("agent://{}/identity/", agent_id); + + // Check for agent_name identity signal + let agent_name_uri = format!("{}agent-name", identity_prefix); + if let Ok(Some(entry)) = VikingStorage::get(storage.as_ref(), &agent_name_uri).await { + // Extract name from content like "助手的名字是小马" + let name = entry.content.strip_prefix("助手的名字是") + .map(|n| n.trim().to_string()) + .unwrap_or_else(|| entry.content.clone()); + + if !name.is_empty() { + // Update IdentityFiles.soul to include the agent name + let mut manager = identity_state.lock().await; + let current_soul = manager.get_file(agent_id, crate::intelligence::identity::IdentityFile::Soul); + + // Only update if the name isn't already in the soul + if !current_soul.contains(&name) { + let updated_soul = if current_soul.is_empty() { + format!("# ZCLAW 人格\n\n你的名字是{}。\n\n你是一个成长性的中文 AI 助手。", name) + } else if current_soul.contains("你的名字是") || current_soul.contains("你的名字：") { + // Replace existing name line + let re = regex::Regex::new(r"你的名字是[^\n]+").unwrap(); + re.replace(¤t_soul, format!("你的名字是{}", name)).to_string() + } else { + // Prepend name to existing soul + format!("你的名字是{}。\n\n{}", name, current_soul) + }; + + if let Err(e) = manager.update_file(agent_id, "soul", &updated_soul) { + warn!("[intelligence_hooks] Failed to update soul with agent name: {}", e); + } else { + debug!("[intelligence_hooks] Updated agent name to '{}' in soul", name); + } + } + drop(manager); + + // Emit event for frontend to update AgentConfig.name + let _ = app.emit("zclaw:agent-identity-updated", serde_json::json!({ + "agentId": agent_id, + "agentName": name, + })); + } + } + + // Check for user_name identity signal + let user_name_uri = format!("{}user-name", identity_prefix); + if let Ok(Some(entry)) = VikingStorage::get(storage.as_ref(), &user_name_uri).await { + let name = entry.content.strip_prefix("用户的名字是") + .map(|n| n.trim().to_string()) + .unwrap_or_else(|| entry.content.clone()); + + if !name.is_empty() { + let mut manager = identity_state.lock().await; + let profile = manager.get_file(agent_id, crate::intelligence::identity::IdentityFile::UserProfile); + + if !profile.contains(&name) { + manager.append_to_user_profile(agent_id, &format!("- 用户名字: {}", name)); + debug!("[intelligence_hooks] Appended user name '{}' to profile", name); + } + } + } + } } /// Build memory context by searching VikingStorage for relevant memories diff --git a/desktop/src-tauri/src/kernel_commands/chat.rs b/desktop/src-tauri/src/kernel_commands/chat.rs index 6cc2a46..700f85a 100644 --- a/desktop/src-tauri/src/kernel_commands/chat.rs +++ b/desktop/src-tauri/src/kernel_commands/chat.rs @@ -324,6 +324,7 @@ pub async fn agent_chat_stream( let hb_state = heartbeat_state.inner().clone(); let rf_state = reflection_state.inner().clone(); + let id_state_hook = identity_state.inner().clone(); // Clone the guard map for cleanup in the spawned task let guard_map: SessionStreamGuard = stream_guard.inner().clone(); @@ -380,12 +381,14 @@ pub async fn agent_chat_stream( let hb = hb_state.clone(); let rf = rf_state.clone(); let driver = llm_driver.clone(); + let id_state = id_state_hook.clone(); + let app_hook = app.clone(); if driver.is_none() { tracing::debug!("[agent_chat_stream] Post-hook firing without LLM driver (schedule intercept path)"); } tokio::spawn(async move { crate::intelligence_hooks::post_conversation_hook( - &agent_id_hook, &message_hook, &hb, &rf, driver, + &agent_id_hook, &message_hook, &hb, &rf, driver, &id_state, &app_hook, ).await; }); } diff --git a/desktop/src/lib/cold-start-mapper.ts b/desktop/src/lib/cold-start-mapper.ts index 112ae71..4e6b1e7 100644 --- a/desktop/src/lib/cold-start-mapper.ts +++ b/desktop/src/lib/cold-start-mapper.ts @@ -146,6 +146,32 @@ export function detectNameSuggestion(message: string): string | undefined { return undefined; } +/** + * Detect if user gives the agent a name (e.g., "叫你小马", "以后叫你小马", "你的名字是小马"). + * Returns the detected agent name or undefined. + */ +export function detectAgentNameSuggestion(message: string): string | undefined { + if (!message) return undefined; + const patterns = [ + /叫你[""''「」]?(\S{1,8})[""''「」]?[吧。！]?/, + /你的名字[是为][""''「」]?(\S{1,8})[""''「」]?[。！]?/, + /以后叫你[""''「」]?(\S{1,8})[""''「」]?[吧。！]?/, + /给你起[个]?名[字]?(?:叫)?[""''「」]?(\S{1,8})[""''「」]?/, + /name you (\S{1,15})/i, + /call you (\S{1,15})/i, + ]; + for (const pattern of patterns) { + const match = message.match(pattern); + if (match && match[1]) { + const name = match[1].replace(/[吧。！，、]/g, '').trim(); + if (name.length >= 1 && name.length <= 8) { + return name; + } + } + } + return undefined; +} + /** * Determine the next cold start phase based on current phase and user message. */ diff --git a/desktop/src/store/chat/streamStore.ts b/desktop/src/store/chat/streamStore.ts index c50f006..f6d9c9d 100644 --- a/desktop/src/store/chat/streamStore.ts +++ b/desktop/src/store/chat/streamStore.ts @@ -34,6 +34,8 @@ import { } from './conversationStore'; import { useMessageStore } from './messageStore'; import { useArtifactStore } from './artifactStore'; +import { llmSuggest } from '../../lib/llm-service'; +import { detectNameSuggestion, detectAgentNameSuggestion } from '../../lib/cold-start-mapper'; const log = createLogger('StreamStore'); @@ -371,7 +373,30 @@ function createCompleteHandler( .map(m => ({ role: m.role, content: m.content })); const convId = useConversationStore.getState().currentConversationId; getMemoryExtractor().extractFromConversation(filtered, agentId, convId ?? undefined) - .then(() => { + .then(async () => { + // Detect name preference from last user message (e.g. "叫我小马") + const lastUserMsg = [...msgs].reverse().find(m => m.role === 'user'); + const detectedName = lastUserMsg ? detectNameSuggestion(lastUserMsg.content) : undefined; + if (detectedName && agentId) { + try { + const { useAgentStore } = await import('../agentStore'); + await useAgentStore.getState().updateClone(agentId, { userName: detectedName }); + log.info(`Updated userName to "${detectedName}" from conversation`); + } catch (e) { + log.warn('Failed to persist detected userName:', e); + } + } + // Detect agent name change (e.g. "叫你小马", "以后叫你小马") + const detectedAgentName = lastUserMsg ? detectAgentNameSuggestion(lastUserMsg.content) : undefined; + if (detectedAgentName && agentId) { + try { + const { useAgentStore } = await import('../agentStore'); + await useAgentStore.getState().updateClone(agentId, { name: detectedAgentName }); + log.info(`Updated agent name to "${detectedAgentName}" from conversation`); + } catch (e) { + log.warn('Failed to persist detected agent name:', e); + } + } if (typeof window !== 'undefined') { window.dispatchEvent(new CustomEvent('zclaw:agent-profile-updated', { detail: { agentId } @@ -391,15 +416,17 @@ function createCompleteHandler( } }); - // Follow-up suggestions + // Follow-up suggestions (async LLM call with keyword fallback) const latestMsgs = chat.getMessages() || []; - const completedMsg = latestMsgs.find(m => m.id === assistantId); - if (completedMsg?.content) { - const suggestions = generateFollowUpSuggestions(completedMsg.content); - if (suggestions.length > 0) { - set({ suggestions }); - } - } + const conversationMessages = latestMsgs + .filter(m => m.role === 'user' || m.role === 'assistant') + .filter(m => !m.streaming) + .map(m => ({ role: m.role, content: m.content })); + + generateLLMSuggestions(conversationMessages, set).catch(err => { + log.warn('Suggestion generation error:', err); + set({ suggestionsLoading: false }); + }); }; } @@ -410,6 +437,8 @@ export interface StreamState { isLoading: boolean; chatMode: ChatModeType; suggestions: string[]; + /** Whether LLM-generated suggestions are being fetched. */ + suggestionsLoading: boolean; /** Run ID of the currently active stream (null when idle). */ activeRunId: string | null; @@ -425,6 +454,7 @@ export interface StreamState { // Suggestions setSuggestions: (suggestions: string[]) => void; + setSuggestionsLoading: (loading: boolean) => void; // Skill search searchSkills: (query: string) => { @@ -440,7 +470,7 @@ export interface StreamState { // Follow-up suggestion generator // --------------------------------------------------------------------------- -function generateFollowUpSuggestions(content: string): string[] { +function generateKeywordFallback(content: string): string[] { const suggestions: string[] = []; const lower = content.toLowerCase(); @@ -473,6 +503,148 @@ function generateFollowUpSuggestions(content: string): string[] { return suggestions; } +/** + * Parse LLM response into an array of suggestion strings. + * Handles: raw JSON array, markdown-fenced JSON, trailing/leading text. + */ +function parseSuggestionResponse(raw: string): string[] { + let cleaned = raw.trim(); + // Strip markdown code fences + cleaned = cleaned.replace(/^```(?:json)?\s*\n?/i, ''); + cleaned = cleaned.replace(/\n?```\s*$/i, ''); + cleaned = cleaned.trim(); + + // Direct JSON parse + try { + const parsed = JSON.parse(cleaned); + if (Array.isArray(parsed)) { + return parsed + .filter((item): item is string => typeof item === 'string' && item.trim().length > 0) + .slice(0, 3); + } + } catch { /* fall through */ } + + // Extract JSON array from surrounding text + const arrayMatch = cleaned.match(/\[[\s\S]*?\]/); + if (arrayMatch) { + try { + const parsed = JSON.parse(arrayMatch[0]); + if (Array.isArray(parsed)) { + return parsed + .filter((item): item is string => typeof item === 'string' && item.trim().length > 0) + .slice(0, 3); + } + } catch { /* fall through */ } + } + + // Last resort: split by newlines, strip list markers + const lines = cleaned + .split(/\n/) + .map(l => l.replace(/^[-*\d.)\]]+\s*/, '').trim()) + .filter(l => l.length > 0 && l.length < 60); + if (lines.length > 0) { + return lines.slice(0, 3); + } + + return []; +} + +/** + * Generate contextual follow-up suggestions via LLM. + * Routes through SaaS relay or local kernel based on connection mode. + * Falls back to keyword-based approach on any failure. + */ +async function generateLLMSuggestions( + messages: Array<{ role: string; content: string }>, + set: (partial: Partial) => void, +): Promise { + set({ suggestionsLoading: true }); + + try { + const recentMessages = messages.slice(-6); + const context = recentMessages + .map(m => `${m.role === 'user' ? '用户' : '助手'}: ${m.content}`) + .join('\n\n'); + + const connectionMode = typeof localStorage !== 'undefined' + ? localStorage.getItem('zclaw-connection-mode') + : null; + + let raw: string; + + if (connectionMode === 'saas') { + // SaaS relay: use saasClient directly for reliable auth + raw = await llmSuggestViaSaaS(context); + } else { + // Local kernel: use llm-service adapter (GatewayLLMAdapter → agent_chat) + raw = await llmSuggest(context); + } + + const suggestions = parseSuggestionResponse(raw); + + if (suggestions.length > 0) { + set({ suggestions, suggestionsLoading: false }); + } else { + const lastAssistant = messages.filter(m => m.role === 'assistant').pop()?.content || ''; + set({ suggestions: generateKeywordFallback(lastAssistant), suggestionsLoading: false }); + } + } catch (err) { + log.warn('LLM suggestion generation failed, using keyword fallback:', err); + const lastAssistant = messages.filter(m => m.role === 'assistant').pop()?.content || ''; + set({ suggestions: generateKeywordFallback(lastAssistant), suggestionsLoading: false }); + } +} + +/** + * Generate suggestions via SaaS relay, using saasStore auth directly. + */ +async function llmSuggestViaSaaS(context: string): Promise { + const { useSaaSStore } = await import('../saasStore'); + const { saasUrl, authToken } = useSaaSStore.getState(); + + if (!saasUrl || !authToken) { + throw new Error('SaaS not authenticated'); + } + + const { saasClient } = await import('../../lib/saas-client'); + saasClient.setBaseUrl(saasUrl); + saasClient.setToken(authToken); + + const response = await saasClient.chatCompletion( + { + model: 'default', + messages: [ + { role: 'system', content: LLM_PROMPTS_SYSTEM }, + { role: 'user', content: `以下是对话中最近的消息：\n\n${context}\n\n请生成 3 个后续问题。` }, + ], + max_tokens: 500, + temperature: 0.7, + stream: false, + }, + AbortSignal.timeout(15000), + ); + + if (!response.ok) { + const errText = await response.text().catch(() => 'unknown error'); + throw new Error(`SaaS relay error ${response.status}: ${errText.substring(0, 100)}`); + } + + const data = await response.json(); + return data?.choices?.[0]?.message?.content || ''; +} + +const LLM_PROMPTS_SYSTEM = `你是对话分析助手。根据最近的对话内容，生成 3 个用户可能想继续探讨的问题。 + +要求： +- 每个问题必须与对话内容直接相关，具体且有针对性 +- 帮助用户深入理解、实际操作或拓展思路 +- 每个问题不超过 30 个中文字符 +- 不要重复对话中已讨论过的内容 +- 使用与用户相同的语言 + +只输出 JSON 数组，包含恰好 3 个字符串。不要输出任何其他内容。 +示例：["如何在生产环境中部署？", "这个方案的成本如何？", "有没有更简单的替代方案？"]`; + // --------------------------------------------------------------------------- // ChatStore injection (avoids circular imports) // --------------------------------------------------------------------------- @@ -499,6 +671,7 @@ export const useStreamStore = create()( isLoading: false, chatMode: 'thinking' as ChatModeType, suggestions: [], + suggestionsLoading: false, activeRunId: null as string | null, // ── Chat Mode ── @@ -508,6 +681,7 @@ export const useStreamStore = create()( getChatModeConfig: () => CHAT_MODES[get().chatMode].config, setSuggestions: (suggestions: string[]) => set({ suggestions }), + setSuggestionsLoading: (loading: boolean) => set({ suggestionsLoading: loading }), setIsLoading: (loading: boolean) => set({ isLoading: loading }), @@ -535,7 +709,7 @@ export const useStreamStore = create()( const currentAgent = convStore.currentAgent; const sessionKey = convStore.sessionKey; - set({ suggestions: [] }); + set({ suggestions: [], suggestionsLoading: false }); const effectiveSessionKey = sessionKey || crypto.randomUUID(); const effectiveAgentId = resolveGatewayAgentId(currentAgent); const agentId = currentAgent?.id || 'zclaw-main'; @@ -849,13 +1023,15 @@ export const useStreamStore = create()( } const latestMsgs = _chat?.getMessages() || []; - const completedMsg = latestMsgs.find(m => m.id === streamingMsg.id); - if (completedMsg?.content) { - const suggestions = generateFollowUpSuggestions(completedMsg.content); - if (suggestions.length > 0) { - get().setSuggestions(suggestions); - } - } + const conversationMessages = latestMsgs + .filter(m => m.role === 'user' || m.role === 'assistant') + .filter(m => !m.streaming) + .map(m => ({ role: m.role, content: m.content })); + + generateLLMSuggestions(conversationMessages, set).catch(err => { + log.warn('Suggestion generation error:', err); + set({ suggestionsLoading: false }); + }); } } } else if (delta.stream === 'hand') {