mcp-probe-kit 3.0.21 → 3.0.23

package/README.md

@@ -15,7 +15,7 @@
 
 > **Talk is cheap, show me the Context.**
 > 
-> mcp-probe-kit is a protocol-level toolkit designed for developers who want AI to truly understand their project's intent. It's not just a collection of 29 tools—it's a context-aware system that helps AI agents grasp what you're building.
+> mcp-probe-kit is a protocol-level toolkit designed for developers who want AI to truly understand their project's intent. It's not just a collection of 27 tools—it's a context-aware system that helps AI agents grasp what you're building.
 
 **Languages**: [English](README.md) | [简体中文](i18n/README.zh-CN.md) | [日本語](i18n/README.ja-JP.md) | [한국어](i18n/README.ko-KR.md) | [Español](i18n/README.es-ES.md) | [Français](i18n/README.fr-FR.md) | [Deutsch](i18n/README.de-DE.md) | [Português (BR)](i18n/README.pt-BR.md)
 
@@ -26,7 +26,7 @@
 
 > 🚀 AI-Powered Complete Development Toolkit - Covering the Entire Development Lifecycle
 
-A powerful MCP (Model Context Protocol) server providing **29 tools** covering the complete workflow from product analysis to final release (Requirements → Design → Development → Quality → Release), all tools support **structured output**.
+A powerful MCP (Model Context Protocol) server providing **27 tools** covering the complete workflow from product analysis to final release (Requirements → Design → Development → Quality → Release), all tools support **structured output**.
 
 **🎉 v3.0 Major Update**: Streamlined tool count, focus on core competencies, eliminate choice paralysis, let AI do more native work
 
@@ -42,7 +42,7 @@ A powerful MCP (Model Context Protocol) server providing **29 tools** covering t
 
 - [Quick Start](https://mcp-probe-kit.bytezonex.com/pages/getting-started.html) - Setup in 5 minutes
 - [Local Memory Stack (Qdrant + Nomic Embed)](docs/memory-local-setup.md) - Docker Compose, ports `50008` / `50012`, MCP env
-- [All Tools](https://mcp-probe-kit.bytezonex.com/pages/all-tools.html) - Complete list of 29 tools
+- [All Tools](https://mcp-probe-kit.bytezonex.com/pages/all-tools.html) - Complete list of 27 tools
 - [Best Practices](https://mcp-probe-kit.bytezonex.com/pages/examples.html) - Full development workflow guide
 - [v3.0 Migration Guide](https://mcp-probe-kit.bytezonex.com/pages/migration.html) - Upgrade from v2.x to v3.0
 
@@ -64,8 +64,8 @@ A powerful MCP (Model Context Protocol) server providing **29 tools** covering t
   - `init_project`, `init_project_context`, `add_feature`, `estimate`, `interview`, `ask_user`
 - **🎨 UI/UX Utilities** (3 tools) - Design systems and UI data synchronization
   - `ui_design_system`, `ui_search`, `sync_ui_data`
-- **🧠 Memory & Cursor History** (7 tools) - Reusable asset memory and local Cursor conversation retrieval
-  - `search_memory`, `read_memory_asset`, `memorize_asset`, `scan_and_extract_patterns`, `cursor_list_conversations`, `cursor_search_conversations`, `cursor_read_conversation`
+- **🧠 Memory & Cursor History** (5 tools) - Reusable asset memory and local Cursor conversation retrieval
+  - `search_memory`, `read_memory_asset`, `memorize_asset`, `scan_and_extract_patterns`, `cursor_read_conversation`
 
 ### 🧠 Code Graph Bridge (GitNexus)
 
@@ -93,14 +93,14 @@ A powerful MCP (Model Context Protocol) server providing **29 tools** covering t
 - Embedding service supports two modes:
   - `ollama`
   - `openai-compatible`
-- Cursor history tools read the local Cursor database directly through Node.js, without Python bridge
+- `cursor_read_conversation` reads the local Cursor database directly through Node.js, without Python bridge
 - Cursor history currently supports:
   - Windows: `%APPDATA%\\Cursor\\User\\globalStorage\\state.vscdb`
   - macOS: `~/Library/Application Support/Cursor/User/globalStorage/state.vscdb`
   - Linux: `~/.config/Cursor/User/globalStorage/state.vscdb`
 
 **Memory tools:**
-- `search_memory` - Semantic search across the shared memory pool (optionally prefer `type` / `tags`)
+- `search_memory` - Semantic search across the shared memory pool (optionally prefer `type` / `tags`); returns `id`, `score`, `summary`, and `description` in both text output and `structuredContent` (for MCP clients that only surface `content[0].text`)
 - `memorize_asset` - Persist reusable code/spec/pattern assets into vector memory
 - `read_memory_asset` - Read full asset content by `asset_id`
 - `scan_and_extract_patterns` - Extract reusable patterns from code/file/directory before deciding whether to persist
@@ -194,8 +194,6 @@ ollama pull nomic-embed-text
 ```
 
 **Cursor history tools:**
-- `cursor_list_conversations` - List recent local Cursor conversations by title/workspace
-- `cursor_search_conversations` - Search local Cursor history by keyword or request id
 - `cursor_read_conversation` - Read a single local Cursor conversation timeline by `composer_id`
 
 ### 🎯 Structured Output
@@ -561,7 +559,7 @@ ollama pull nomic-embed-text
 
 ### Cursor History Support
 
-Cursor local history tools do not require Qdrant or embedding configuration.
+`cursor_read_conversation` does not require Qdrant or embedding configuration.
 
 Supported platforms:
 

package/build/index.js

@@ -5,7 +5,7 @@ import { InMemoryTaskMessageQueue, InMemoryTaskStore, } from "@modelcontextproto
 import { CallToolRequestSchema, ListToolsRequestSchema, ListResourcesRequestSchema, ProgressNotificationSchema, ReadResourceRequestSchema, } from "@modelcontextprotocol/sdk/types.js";
 import * as fs from "node:fs";
 import * as path from "node:path";
-import { initProject, gencommit, codeReview, codeInsight, gentest, refactor, initProjectContext, addFeature, fixBug, estimate, startFeature, startBugfix, startOnboard, startRalph, interview, askUser, uiDesignSystem, uiSearch, syncUiData, startUi, startProduct, gitWorkReport, searchMemory, readMemoryAsset, memorizeAsset, scanAndExtractPatterns, cursorListConversations, cursorSearchConversations, cursorReadConversation } from "./tools/index.js";
+import { initProject, gencommit, codeReview, codeInsight, gentest, refactor, initProjectContext, addFeature, fixBug, estimate, startFeature, startBugfix, startOnboard, startRalph, interview, askUser, uiDesignSystem, uiSearch, syncUiData, startUi, startProduct, gitWorkReport, searchMemory, readMemoryAsset, memorizeAsset, scanAndExtractPatterns, cursorReadConversation } from "./tools/index.js";
 import { VERSION, NAME } from "./version.js";
 import { allToolSchemas } from "./schemas/index.js";
 import { filterTools, getToolsetFromEnv } from "./lib/toolset-manager.js";
@@ -467,10 +467,6 @@ async function executeTool(name, args, context) {
             return await memorizeAsset(args);
         case "scan_and_extract_patterns":
             return await scanAndExtractPatterns(args);
-        case "cursor_list_conversations":
-            return await cursorListConversations(args);
-        case "cursor_search_conversations":
-            return await cursorSearchConversations(args);
         case "cursor_read_conversation":
             return await cursorReadConversation(args);
         default:

package/build/lib/memory-orchestration.d.ts

@@ -13,6 +13,7 @@ export interface MemoryInjectionContext {
 }
 export declare function truncateInjectionText(value: string, maxChars: number): string;
 export declare function loadMemoryInjectionContext(query: string, kind?: MemoryPlanKind): Promise<MemoryInjectionContext>;
+export declare function formatSearchMemoryResultsText(results: MemorySearchResult[], config?: MemoryConfig): string;
 export declare function shouldShowSourceInSearch(item: MemorySearchResult, config?: MemoryConfig): boolean;
 export declare function renderMemoryGuideSection(context: MemoryInjectionContext): string;
 export declare function buildMemoryPlanStep(kind?: MemoryPlanKind): {

package/build/lib/memory-orchestration.js

@@ -77,6 +77,27 @@ function formatMemoryResultLabel(item) {
             : '历史资产';
     return `${item.name} [${item.type}] (${kind})`;
 }
+export function formatSearchMemoryResultsText(results, config = getMemoryConfig()) {
+    if (results.length === 0) {
+        return '未找到相关记忆';
+    }
+    const header = `找到 ${results.length} 条相关记忆`;
+    const items = results.map((item, index) => {
+        const lines = [
+            `${index + 1}. ${item.name} [${item.type}] score=${item.score.toFixed(3)}`,
+            `   - id: ${item.id}`,
+            item.summary ? `   - 摘要: ${item.summary}` : '',
+            item.description ? `   - 描述: ${item.description}` : '',
+            item.tags.length > 0 ? `   - 标签: ${item.tags.join(', ')}` : '',
+        ];
+        if (shouldShowSourceInSearch(item, config) && item.sourcePath) {
+            lines.push(`   - 来源: ${item.sourcePath}`);
+        }
+        lines.push(`   - 全文: read_memory_asset {"asset_id": "${item.id}"}`);
+        return lines.filter(Boolean).join('\n');
+    });
+    return `${header}\n\n${items.join('\n\n')}`;
+}
 export function shouldShowSourceInSearch(item, config = getMemoryConfig()) {
     if (config.searchShowSource) {
         return Boolean(item.sourcePath);

package/build/lib/shadcn-ui.d.ts

@@ -0,0 +1,11 @@
+/**
+ * 判断是否适合使用 shadcn/ui 实现路径
+ */
+export declare function isShadcnStack(stack?: string): boolean;
+export declare function isShadcnCategory(category?: string): boolean;
+export declare function isThemeCategory(category?: string): boolean;
+export declare function isGuidelineCategory(category?: string): boolean;
+export declare function formatShadcnResult(data: Record<string, any>): string;
+export declare function formatThemeResult(data: Record<string, any>): string;
+export declare function formatGuidelineResult(data: Record<string, any>): string;
+export declare function pickThemeForProductType(themes: Record<string, any>[], productType: string): Record<string, any> | undefined;

package/build/lib/shadcn-ui.js

@@ -0,0 +1,78 @@
+/**
+ * 判断是否适合使用 shadcn/ui 实现路径
+ */
+export function isShadcnStack(stack) {
+    if (!stack)
+        return false;
+    const normalized = stack.toLowerCase();
+    return (normalized.includes('react') ||
+        normalized.includes('next') ||
+        normalized.includes('shadcn') ||
+        normalized.includes('tailwind'));
+}
+export function isShadcnCategory(category) {
+    return Boolean(category?.startsWith('shadcn-'));
+}
+export function isThemeCategory(category) {
+    return category === 'ui-themes';
+}
+export function isGuidelineCategory(category) {
+    return category === 'ui-guidelines-vercel';
+}
+export function formatShadcnResult(data) {
+    const lines = [
+        `- **名称**: ${data.name || data.title}`,
+        `- **类型**: ${data.type || 'unknown'}`,
+        `- **描述**: ${data.description || '—'}`,
+        `- **安装**: \`${data.installCommand || `npx shadcn@latest add ${data.name}`}\``,
+    ];
+    if (Array.isArray(data.registryDependencies) && data.registryDependencies.length > 0) {
+        lines.push(`- **依赖组件**: ${data.registryDependencies.join(', ')}`);
+    }
+    if (Array.isArray(data.files) && data.files.length > 0) {
+        lines.push(`- **文件**: ${data.files.slice(0, 5).join(', ')}${data.files.length > 5 ? '…' : ''}`);
+    }
+    return lines.join('\n');
+}
+export function formatThemeResult(data) {
+    const lines = [
+        `- **主题**: ${data.title || data.name}`,
+        `- **描述**: ${data.description || '—'}`,
+        `- **基色**: ${data.baseColor || '—'}`,
+        `- **适合**: ${Array.isArray(data.bestFor) ? data.bestFor.join(', ') : '—'}`,
+        `- **用法**: 将 \`globalsCssSnippet\` 粘贴到 \`app/globals.css\`（shadcn new-york）`,
+    ];
+    return lines.join('\n');
+}
+export function formatGuidelineResult(data) {
+    return [
+        `- **级别**: ${data.level || '—'}`,
+        `- **章节**: ${data.section || '—'} / ${data.subsection || '—'}`,
+        `- **规则**: ${data.rule || data.description || '—'}`,
+        `- **来源**: Vercel Web Interface Guidelines`,
+    ].join('\n');
+}
+export function pickThemeForProductType(themes, productType) {
+    if (themes.length === 0)
+        return undefined;
+    const normalized = productType.toLowerCase();
+    const scored = themes.map((theme) => {
+        const bestFor = Array.isArray(theme.bestFor) ? theme.bestFor : [];
+        const score = bestFor.reduce((acc, item) => {
+            const token = item.toLowerCase();
+            if (normalized.includes(token) || token.includes(normalized)) {
+                return acc + 2;
+            }
+            if (normalized.split(/\s+/).some((part) => token.includes(part) && part.length > 2)) {
+                return acc + 1;
+            }
+            return acc;
+        }, 0);
+        return { theme, score };
+    });
+    scored.sort((a, b) => b.score - a.score);
+    if (scored[0]?.score > 0) {
+        return scored[0].theme;
+    }
+    return themes.find((theme) => theme.name === 'zinc-neutral') || themes[0];
+}