luma-mcp 1.0.2 → 1.1.0

package/build/siliconflow-client.js

@@ -0,0 +1,85 @@
+/**
+ * 硅基流动 DeepSeek-OCR API 客户端
+ * 基于 OpenAI 兼容 API
+ */
+import axios from 'axios';
+import { logger } from './utils/logger.js';
+/**
+ * 硅基流动 API 客户端
+ */
+export class SiliconFlowClient {
+    config;
+    apiEndpoint = 'https://api.siliconflow.cn/v1/chat/completions';
+    constructor(config) {
+        this.config = config;
+    }
+    /**
+     * 分析图片
+     */
+    async analyzeImage(imageDataUrl, prompt, enableThinking) {
+        const requestBody = {
+            model: this.config.model,
+            messages: [
+                {
+                    role: 'user',
+                    content: [
+                        {
+                            type: 'image_url',
+                            image_url: {
+                                url: imageDataUrl,
+                            },
+                        },
+                        {
+                            type: 'text',
+                            text: prompt,
+                        },
+                    ],
+                },
+            ],
+            temperature: this.config.temperature,
+            max_tokens: this.config.maxTokens,
+            top_p: this.config.topP,
+            stream: false,
+        };
+        logger.info('Calling SiliconFlow DeepSeek-OCR API', {
+            model: this.config.model,
+        });
+        try {
+            const response = await axios.post(this.apiEndpoint, requestBody, {
+                headers: {
+                    'Authorization': `Bearer ${this.config.apiKey}`,
+                    'Content-Type': 'application/json',
+                },
+                timeout: 60000, // 60秒超时
+            });
+            if (!response.data.choices || response.data.choices.length === 0) {
+                throw new Error('No response from DeepSeek-OCR');
+            }
+            const result = response.data.choices[0].message.content;
+            const usage = response.data.usage;
+            logger.info('SiliconFlow API call successful', {
+                tokens: usage?.total_tokens || 0,
+                model: response.data.model
+            });
+            return result;
+        }
+        catch (error) {
+            logger.error('SiliconFlow API call failed', {
+                error: error instanceof Error ? error.message : String(error)
+            });
+            if (axios.isAxiosError(error)) {
+                const message = error.response?.data?.error?.message || error.message;
+                const status = error.response?.status;
+                throw new Error(`SiliconFlow API error (${status || 'unknown'}): ${message}`);
+            }
+            throw error;
+        }
+    }
+    /**
+     * 获取模型名称
+     */
+    getModelName() {
+        return this.config.model;
+    }
+}
+//# sourceMappingURL=siliconflow-client.js.map

package/build/siliconflow-client.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"siliconflow-client.js","sourceRoot":"","sources":["../src/siliconflow-client.ts"],"names":[],"mappings":"AAAA;;;GAGG;AAEH,OAAO,KAAK,MAAM,OAAO,CAAC;AAG1B,OAAO,EAAE,MAAM,EAAE,MAAM,mBAAmB,CAAC;AA0C3C;;GAEG;AACH,MAAM,OAAO,iBAAiB;IACpB,MAAM,CAAa;IACnB,WAAW,GAAG,gDAAgD,CAAC;IAEvE,YAAY,MAAkB;QAC5B,IAAI,CAAC,MAAM,GAAG,MAAM,CAAC;IACvB,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,YAAY,CAAC,YAAoB,EAAE,MAAc,EAAE,cAAwB;QAC/E,MAAM,WAAW,GAAuB;YACtC,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,KAAK;YACxB,QAAQ,EAAE;gBACR;oBACE,IAAI,EAAE,MAAM;oBACZ,OAAO,EAAE;wBACP;4BACE,IAAI,EAAE,WAAW;4BACjB,SAAS,EAAE;gCACT,GAAG,EAAE,YAAY;6BAClB;yBACF;wBACD;4BACE,IAAI,EAAE,MAAM;4BACZ,IAAI,EAAE,MAAM;yBACb;qBACF;iBACF;aACF;YACD,WAAW,EAAE,IAAI,CAAC,MAAM,CAAC,WAAW;YACpC,UAAU,EAAE,IAAI,CAAC,MAAM,CAAC,SAAS;YACjC,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,IAAI;YACvB,MAAM,EAAE,KAAK;SACd,CAAC;QAEF,MAAM,CAAC,IAAI,CAAC,sCAAsC,EAAE;YAClD,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,KAAK;SACzB,CAAC,CAAC;QAEH,IAAI,CAAC;YACH,MAAM,QAAQ,GAAG,MAAM,KAAK,CAAC,IAAI,CAC/B,IAAI,CAAC,WAAW,EAChB,WAAW,EACX;gBACE,OAAO,EAAE;oBACP,eAAe,EAAE,UAAU,IAAI,CAAC,MAAM,CAAC,MAAM,EAAE;oBAC/C,cAAc,EAAE,kBAAkB;iBACnC;gBACD,OAAO,EAAE,KAAK,EAAE,QAAQ;aACzB,CACF,CAAC;YAEF,IAAI,CAAC,QAAQ,CAAC,IAAI,CAAC,OAAO,IAAI,QAAQ,CAAC,IAAI,CAAC,OAAO,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;gBACjE,MAAM,IAAI,KAAK,CAAC,+BAA+B,CAAC,CAAC;YACnD,CAAC;YAED,MAAM,MAAM,GAAG,QAAQ,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,OAAO,CAAC;YACxD,MAAM,KAAK,GAAG,QAAQ,CAAC,IAAI,CAAC,KAAK,CAAC;YAElC,MAAM,CAAC,IAAI,CAAC,iCAAiC,EAAE;gBAC7C,MAAM,EAAE,KAAK,EAAE,YAAY,IAAI,CAAC;gBAChC,KAAK,EAAE,QAAQ,CAAC,IAAI,CAAC,KAAK;aAC3B,CAAC,CAAC;YAEH,OAAO,MAAM,CAAC;QAChB,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,MAAM,CAAC,KAAK,CAAC,6BAA6B,EAAE;gBAC1C,KAAK,EAAE,KAAK,YAAY,KAAK,CAAC,CAAC,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC;aAC9D,CAAC,CAAC;YAEH,IAAI,KAAK,CAAC,YAAY,CAAC,KAAK,CAAC,EAAE,CAAC;gBAC9B,MAAM,OAAO,GAAG,KAAK,CAAC,QAAQ,EAAE,IAAI,EAAE,KAAK,EAAE,OAAO,IAAI,KAAK,CAAC,OAAO,CAAC;gBACtE,MAAM,MAAM,GAAG,KAAK,CAAC,QAAQ,EAAE,MAAM,CAAC;gBACtC,MAAM,IAAI,KAAK,CAAC,0BAA0B,MAAM,IAAI,SAAS,MAAM,OAAO,EAAE,CAAC,CAAC;YAChF,CAAC;YACD,MAAM,KAAK,CAAC;QACd,CAAC;IACH,CAAC;IAED;;OAEG;IACH,YAAY;QACV,OAAO,IAAI,CAAC,MAAM,CAAC,KAAK,CAAC;IAC3B,CAAC;CACF"}

package/build/vision-client.d.ts

@@ -0,0 +1,18 @@
+/**
+ * 视觉模型客户端统一接口
+ */
+export interface VisionClient {
+    /**
+     * 分析图片
+     * @param imageDataUrl 图片 Data URL 或 URL
+     * @param prompt 分析提示词
+     * @param enableThinking 是否启用思考模式（如果模型支持）
+     * @returns 分析结果文本
+     */
+    analyzeImage(imageDataUrl: string, prompt: string, enableThinking?: boolean): Promise<string>;
+    /**
+     * 获取模型名称
+     */
+    getModelName(): string;
+}
+//# sourceMappingURL=vision-client.d.ts.map

package/build/vision-client.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"vision-client.d.ts","sourceRoot":"","sources":["../src/vision-client.ts"],"names":[],"mappings":"AAAA;;GAEG;AAEH,MAAM,WAAW,YAAY;IAC3B;;;;;;OAMG;IACH,YAAY,CAAC,YAAY,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,cAAc,CAAC,EAAE,OAAO,GAAG,OAAO,CAAC,MAAM,CAAC,CAAC;IAE9F;;OAEG;IACH,YAAY,IAAI,MAAM,CAAC;CACxB"}

package/build/vision-client.js

@@ -0,0 +1,5 @@
+/**
+ * 视觉模型客户端统一接口
+ */
+export {};
+//# sourceMappingURL=vision-client.js.map

package/build/vision-client.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"vision-client.js","sourceRoot":"","sources":["../src/vision-client.ts"],"names":[],"mappings":"AAAA;;GAEG"}

package/build/zhipu-client.d.ts

@@ -2,10 +2,11 @@
  * 智谱 GLM-4.5V API 客户端
  */
 import type { LumaConfig } from './config.js';
+import type { VisionClient } from './vision-client.js';
 /**
  * 智谱 API 客户端
  */
-export declare class ZhipuClient {
+export declare class ZhipuClient implements VisionClient {
     private config;
     private apiEndpoint;
     constructor(config: LumaConfig);
@@ -13,5 +14,9 @@ export declare class ZhipuClient {
      * 分析图片
      */
     analyzeImage(imageDataUrl: string, prompt: string, enableThinking?: boolean): Promise<string>;
+    /**
+     * 获取模型名称
+     */
+    getModelName(): string;
 }
 //# sourceMappingURL=zhipu-client.d.ts.map

package/build/zhipu-client.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"zhipu-client.d.ts","sourceRoot":"","sources":["../src/zhipu-client.ts"],"names":[],"mappings":"AAAA;;GAEG;AAGH,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,aAAa,CAAC;AA4C9C;;GAEG;AACH,qBAAa,WAAW;IACtB,OAAO,CAAC,MAAM,CAAa;IAC3B,OAAO,CAAC,WAAW,CAA2D;gBAElE,MAAM,EAAE,UAAU;IAI9B;;OAEG;IACG,YAAY,CAAC,YAAY,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,cAAc,CAAC,EAAE,OAAO,GAAG,OAAO,CAAC,MAAM,CAAC;CA0EpG"}
+{"version":3,"file":"zhipu-client.d.ts","sourceRoot":"","sources":["../src/zhipu-client.ts"],"names":[],"mappings":"AAAA;;GAEG;AAGH,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,aAAa,CAAC;AAC9C,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,oBAAoB,CAAC;AA4CvD;;GAEG;AACH,qBAAa,WAAY,YAAW,YAAY;IAC9C,OAAO,CAAC,MAAM,CAAa;IAC3B,OAAO,CAAC,WAAW,CAA2D;gBAElE,MAAM,EAAE,UAAU;IAI9B;;OAEG;IACG,YAAY,CAAC,YAAY,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,cAAc,CAAC,EAAE,OAAO,GAAG,OAAO,CAAC,MAAM,CAAC;IA4EnG;;OAEG;IACH,YAAY,IAAI,MAAM;CAGvB"}

package/build/zhipu-client.js

@@ -38,10 +38,11 @@ export class ZhipuClient {
             temperature: this.config.temperature,
             max_tokens: this.config.maxTokens,
             top_p: this.config.topP,
+            thinking: { type: 'enabled' }, // 默认启用思考模式，提高分析准确性
         };
-        // 允许显式启用 thinking（默认关闭以提高速度和降低成本）
-        if (this.config.enableThinking || enableThinking === true) {
-            requestBody.thinking = { type: 'enabled' };
+        // 允许显式禁用 thinking（如需要更快速度）
+        if (this.config.enableThinking === false || enableThinking === false) {
+            delete requestBody.thinking;
         }
         logger.info('Calling GLM-4.5V API', {
             model: this.config.model,
@@ -78,5 +79,11 @@ export class ZhipuClient {
             throw error;
         }
     }
+    /**
+     * 获取模型名称
+     */
+    getModelName() {
+        return this.config.model;
+    }
 }
 //# sourceMappingURL=zhipu-client.js.map

package/build/zhipu-client.js.map

@@ -1 +1 @@
-{"version":3,"file":"zhipu-client.js","sourceRoot":"","sources":["../src/zhipu-client.ts"],"names":[],"mappings":"AAAA;;GAEG;AAEH,OAAO,KAAK,MAAM,OAAO,CAAC;AAE1B,OAAO,EAAE,MAAM,EAAE,MAAM,mBAAmB,CAAC;AA2C3C;;GAEG;AACH,MAAM,OAAO,WAAW;IACd,MAAM,CAAa;IACnB,WAAW,GAAG,uDAAuD,CAAC;IAE9E,YAAY,MAAkB;QAC5B,IAAI,CAAC,MAAM,GAAG,MAAM,CAAC;IACvB,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,YAAY,CAAC,YAAoB,EAAE,MAAc,EAAE,cAAwB;QAC/E,MAAM,WAAW,GAAiB;YAChC,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,KAAK;YACxB,QAAQ,EAAE;gBACR;oBACE,IAAI,EAAE,MAAM;oBACZ,OAAO,EAAE;wBACP;4BACE,IAAI,EAAE,WAAW;4BACjB,SAAS,EAAE;gCACT,GAAG,EAAE,YAAY;6BAClB;yBACF;wBACD;4BACE,IAAI,EAAE,MAAM;4BACZ,IAAI,EAAE,MAAM;yBACb;qBACF;iBACF;aACF;YACD,WAAW,EAAE,IAAI,CAAC,MAAM,CAAC,WAAW;YACpC,UAAU,EAAE,IAAI,CAAC,MAAM,CAAC,SAAS;YACjC,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,IAAI;SACxB,CAAC;QAEF,kCAAkC;QAClC,IAAI,IAAI,CAAC,MAAM,CAAC,cAAc,IAAI,cAAc,KAAK,IAAI,EAAE,CAAC;YAC1D,WAAW,CAAC,QAAQ,GAAG,EAAE,IAAI,EAAE,SAAS,EAAE,CAAC;QAC7C,CAAC;QAED,MAAM,CAAC,IAAI,CAAC,sBAAsB,EAAE;YAClC,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,KAAK;YACxB,QAAQ,EAAE,CAAC,CAAC,WAAW,CAAC,QAAQ;SACjC,CAAC,CAAC;QAEH,IAAI,CAAC;YACH,MAAM,QAAQ,GAAG,MAAM,KAAK,CAAC,IAAI,CAC/B,IAAI,CAAC,WAAW,EAChB,WAAW,EACX;gBACE,OAAO,EAAE;oBACP,eAAe,EAAE,UAAU,IAAI,CAAC,MAAM,CAAC,MAAM,EAAE;oBAC/C,cAAc,EAAE,kBAAkB;iBACnC;gBACD,OAAO,EAAE,KAAK,EAAE,QAAQ;aACzB,CACF,CAAC;YAEF,IAAI,CAAC,QAAQ,CAAC,IAAI,CAAC,OAAO,IAAI,QAAQ,CAAC,IAAI,CAAC,OAAO,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;gBACjE,MAAM,IAAI,KAAK,CAAC,2BAA2B,CAAC,CAAC;YAC/C,CAAC;YAED,MAAM,MAAM,GAAG,QAAQ,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,OAAO,CAAC;YACxD,MAAM,KAAK,GAAG,QAAQ,CAAC,IAAI,CAAC,KAAK,CAAC;YAElC,MAAM,CAAC,IAAI,CAAC,8BAA8B,EAAE;gBAC1C,MAAM,EAAE,KAAK,EAAE,YAAY,IAAI,CAAC;gBAChC,KAAK,EAAE,QAAQ,CAAC,IAAI,CAAC,KAAK;aAC3B,CAAC,CAAC;YAEH,OAAO,MAAM,CAAC;QAChB,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,MAAM,CAAC,KAAK,CAAC,0BAA0B,EAAE;gBACvC,KAAK,EAAE,KAAK,YAAY,KAAK,CAAC,CAAC,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC;aAC9D,CAAC,CAAC;YAEH,IAAI,KAAK,CAAC,YAAY,CAAC,KAAK,CAAC,EAAE,CAAC;gBAC9B,MAAM,OAAO,GAAG,KAAK,CAAC,QAAQ,EAAE,IAAI,EAAE,KAAK,EAAE,OAAO,IAAI,KAAK,CAAC,OAAO,CAAC;gBACtE,MAAM,MAAM,GAAG,KAAK,CAAC,QAAQ,EAAE,MAAM,CAAC;gBACtC,MAAM,IAAI,KAAK,CAAC,uBAAuB,MAAM,IAAI,SAAS,MAAM,OAAO,EAAE,CAAC,CAAC;YAC7E,CAAC;YACD,MAAM,KAAK,CAAC;QACd,CAAC;IACH,CAAC;CACF"}
+{"version":3,"file":"zhipu-client.js","sourceRoot":"","sources":["../src/zhipu-client.ts"],"names":[],"mappings":"AAAA;;GAEG;AAEH,OAAO,KAAK,MAAM,OAAO,CAAC;AAG1B,OAAO,EAAE,MAAM,EAAE,MAAM,mBAAmB,CAAC;AA2C3C;;GAEG;AACH,MAAM,OAAO,WAAW;IACd,MAAM,CAAa;IACnB,WAAW,GAAG,uDAAuD,CAAC;IAE9E,YAAY,MAAkB;QAC5B,IAAI,CAAC,MAAM,GAAG,MAAM,CAAC;IACvB,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,YAAY,CAAC,YAAoB,EAAE,MAAc,EAAE,cAAwB;QAC/E,MAAM,WAAW,GAAiB;YAChC,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,KAAK;YACxB,QAAQ,EAAE;gBACR;oBACE,IAAI,EAAE,MAAM;oBACZ,OAAO,EAAE;wBACP;4BACE,IAAI,EAAE,WAAW;4BACjB,SAAS,EAAE;gCACT,GAAG,EAAE,YAAY;6BAClB;yBACF;wBACD;4BACE,IAAI,EAAE,MAAM;4BACZ,IAAI,EAAE,MAAM;yBACb;qBACF;iBACF;aACF;YACD,WAAW,EAAE,IAAI,CAAC,MAAM,CAAC,WAAW;YACpC,UAAU,EAAE,IAAI,CAAC,MAAM,CAAC,SAAS;YACjC,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,IAAI;YACvB,QAAQ,EAAE,EAAE,IAAI,EAAE,SAAS,EAAE,EAAE,mBAAmB;SACnD,CAAC;QAEF,2BAA2B;QAC3B,IAAI,IAAI,CAAC,MAAM,CAAC,cAAc,KAAK,KAAK,IAAI,cAAc,KAAK,KAAK,EAAE,CAAC;YACrE,OAAO,WAAW,CAAC,QAAQ,CAAC;QAC9B,CAAC;QAED,MAAM,CAAC,IAAI,CAAC,sBAAsB,EAAE;YAClC,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,KAAK;YACxB,QAAQ,EAAE,CAAC,CAAC,WAAW,CAAC,QAAQ;SACjC,CAAC,CAAC;QAEH,IAAI,CAAC;YACH,MAAM,QAAQ,GAAG,MAAM,KAAK,CAAC,IAAI,CAC/B,IAAI,CAAC,WAAW,EAChB,WAAW,EACX;gBACE,OAAO,EAAE;oBACP,eAAe,EAAE,UAAU,IAAI,CAAC,MAAM,CAAC,MAAM,EAAE;oBAC/C,cAAc,EAAE,kBAAkB;iBACnC;gBACD,OAAO,EAAE,KAAK,EAAE,QAAQ;aACzB,CACF,CAAC;YAEF,IAAI,CAAC,QAAQ,CAAC,IAAI,CAAC,OAAO,IAAI,QAAQ,CAAC,IAAI,CAAC,OAAO,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;gBACjE,MAAM,IAAI,KAAK,CAAC,2BAA2B,CAAC,CAAC;YAC/C,CAAC;YAED,MAAM,MAAM,GAAG,QAAQ,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,OAAO,CAAC;YACxD,MAAM,KAAK,GAAG,QAAQ,CAAC,IAAI,CAAC,KAAK,CAAC;YAElC,MAAM,CAAC,IAAI,CAAC,8BAA8B,EAAE;gBAC1C,MAAM,EAAE,KAAK,EAAE,YAAY,IAAI,CAAC;gBAChC,KAAK,EAAE,QAAQ,CAAC,IAAI,CAAC,KAAK;aAC3B,CAAC,CAAC;YAEH,OAAO,MAAM,CAAC;QAChB,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,MAAM,CAAC,KAAK,CAAC,0BAA0B,EAAE;gBACvC,KAAK,EAAE,KAAK,YAAY,KAAK,CAAC,CAAC,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC;aAC9D,CAAC,CAAC;YAEH,IAAI,KAAK,CAAC,YAAY,CAAC,KAAK,CAAC,EAAE,CAAC;gBAC9B,MAAM,OAAO,GAAG,KAAK,CAAC,QAAQ,EAAE,IAAI,EAAE,KAAK,EAAE,OAAO,IAAI,KAAK,CAAC,OAAO,CAAC;gBACtE,MAAM,MAAM,GAAG,KAAK,CAAC,QAAQ,EAAE,MAAM,CAAC;gBACtC,MAAM,IAAI,KAAK,CAAC,uBAAuB,MAAM,IAAI,SAAS,MAAM,OAAO,EAAE,CAAC,CAAC;YAC7E,CAAC;YACD,MAAM,KAAK,CAAC;QACd,CAAC;IACH,CAAC;IAED;;OAEG;IACH,YAAY;QACV,OAAO,IAAI,CAAC,MAAM,CAAC,KAAK,CAAC;IAC3B,CAAC;CACF"}

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "luma-mcp",
-  "version": "1.0.2",
-  "description": "A vision understanding MCP server powered by GLM-4.5V",
+  "version": "1.1.0",
+  "description": "Multi-model vision understanding MCP server. Supports GLM-4.5V (Zhipu) and DeepSeek-OCR (SiliconFlow - Free)",
   "type": "module",
   "bin": {
     "luma-mcp": "build/index.js"
@@ -19,7 +19,12 @@
     "ai",
     "glm-4.5v",
     "zhipu",
-    "image-understanding"
+    "deepseek-ocr",
+    "siliconflow",
+    "ocr",
+    "free",
+    "image-understanding",
+    "multi-model"
   ],
   "author": "Jochen",
   "license": "MIT",

package/test/test-deepseek-raw.ts

@@ -0,0 +1,94 @@
+/**
+ * 直接测试 DeepSeek-OCR API（无任何包装）
+ */
+
+import axios from 'axios';
+import * as fs from 'fs';
+import * as path from 'path';
+
+async function testDeepSeekOCR(imagePath: string) {
+  console.log('\n🧪 测试 DeepSeek-OCR API（原始调用）\n');
+
+  const apiKey = 'sk-skrldwndjawxvzzomztwmoinnwmvumezqyejysqutjwkjcdt';
+  
+  // 读取图片并转为 base64
+  const imageBuffer = fs.readFileSync(imagePath);
+  const base64Image = imageBuffer.toString('base64');
+  const mimeType = imagePath.endsWith('.png') ? 'image/png' : 'image/jpeg';
+  const imageDataUrl = `data:${mimeType};base64,${base64Image}`;
+
+  console.log(`📸 图片: ${imagePath}`);
+  console.log(`📦 大小: ${(imageBuffer.length / 1024).toFixed(2)} KB\n`);
+
+  // 测试不同的 prompt
+  const prompts = [
+    '识别图片中的所有文字',
+    'OCR',
+    'Extract all text from this image',
+    'What do you see in this image?',
+    '请详细描述这张图片'
+  ];
+
+  for (const prompt of prompts) {
+    console.log(`\n🔍 测试 Prompt: "${prompt}"`);
+    console.log('─'.repeat(50));
+
+    try {
+      const response = await axios.post(
+        'https://api.siliconflow.cn/v1/chat/completions',
+        {
+          model: 'deepseek-ai/DeepSeek-OCR',
+          messages: [
+            {
+              role: 'user',
+              content: [
+                {
+                  type: 'image_url',
+                  image_url: {
+                    url: imageDataUrl,
+                  },
+                },
+                {
+                  type: 'text',
+                  text: prompt,
+                },
+              ],
+            },
+          ],
+          temperature: 0.7,
+          max_tokens: 4096,
+        },
+        {
+          headers: {
+            'Authorization': `Bearer ${apiKey}`,
+            'Content-Type': 'application/json',
+          },
+          timeout: 60000,
+        }
+      );
+
+      const result = response.data.choices[0].message.content;
+      const usage = response.data.usage;
+
+      console.log(`✅ Tokens: ${usage.total_tokens} (prompt: ${usage.prompt_tokens}, completion: ${usage.completion_tokens})`);
+      console.log(`📝 响应长度: ${result?.length || 0} 字符`);
+      
+      if (result && result.trim().length > 0) {
+        console.log('\n📊 结果:');
+        console.log('─'.repeat(50));
+        console.log(result);
+        console.log('─'.repeat(50));
+        console.log('\n✅ 找到有效响应！');
+        break;
+      } else {
+        console.log('❌ 空响应');
+      }
+    } catch (error: any) {
+      console.log(`❌ 错误: ${error.message}`);
+    }
+  }
+}
+
+// 运行测试
+const imagePath = path.join(process.cwd(), 'test.png');
+testDeepSeekOCR(imagePath).catch(console.error);

package/test/test-local.ts

@@ -4,7 +4,9 @@
  */
 
 import { loadConfig } from '../src/config.js';
+import type { VisionClient } from '../src/vision-client.js';
 import { ZhipuClient } from '../src/zhipu-client.js';
+import { SiliconFlowClient } from '../src/siliconflow-client.js';
 import { imageToBase64, validateImageSource } from '../src/image-processor.js';
 import { buildAnalysisPrompt } from '../src/prompts.js';
 import { logger } from '../src/utils/logger.js';
@@ -18,7 +20,7 @@ async function testImageAnalysis(imagePath: string, question?: string) {
     // 1. 加载配置
     console.log('📝 加载配置...');
     const config = loadConfig();
-    console.log(`✅ 配置加载成功: 模型 ${config.model}\n`);
+    console.log(`✅ 配置加载成功: 提供商 ${config.provider}, 模型 ${config.model}\n`);
 
     // 2. 验证图片
     console.log('🔍 验证图片来源...');
@@ -33,12 +35,19 @@ async function testImageAnalysis(imagePath: string, question?: string) {
 
     // 4. 构建提示词
     console.log('💬 构建提示词...');
-    const prompt = buildAnalysisPrompt(question);
+    // DeepSeek-OCR 需要简洁 prompt
+    const prompt = config.provider === 'siliconflow'
+      ? (question || '请详细分析这张图片的内容')
+      : buildAnalysisPrompt(question);
     console.log(`✅ 提示词: ${question || '通用描述'}\n`);
 
-    // 5. 调用API
-    console.log('🤖 调用 GLM-4.5V API...');
-    const client = new ZhipuClient(config);
+    // 5. 创建客户端并调用API
+    const client: VisionClient = config.provider === 'siliconflow'
+      ? new SiliconFlowClient(config)
+      : new ZhipuClient(config);
+    
+    const modelName = config.provider === 'siliconflow' ? 'DeepSeek-OCR' : 'GLM-4.5V';
+    console.log(`🤖 调用 ${modelName} API...`);
     const result = await client.analyzeImage(imageDataUrl, prompt);
 
     // 6. 显示结果
@@ -76,8 +85,12 @@ if (args.length === 0) {
   npm run test:local https://example.com/image.jpg
 
 环境变量:
-  ZHIPU_API_KEY=your-api-key  # 必需
-  ZHIPU_MODEL=glm-4.5v        # 可选
+  # 使用智谱 GLM-4.5V
+  ZHIPU_API_KEY=your-api-key
+  
+  # 使用硅基流动 DeepSeek-OCR
+  MODEL_PROVIDER=siliconflow
+  SILICONFLOW_API_KEY=your-api-key
   `);
   process.exit(1);
 }

package/.claude/settings.local.json

@@ -1,10 +0,0 @@
-{
-  "permissions": {
-    "allow": [
-      "mcp__devmind__list_projects",
-      "mcp__time__current_time"
-    ],
-    "deny": [],
-    "ask": []
-  }
-}

package/mcp-server/README.md

@@ -1,41 +0,0 @@
-# ZAI MCP Server
-
-[中文文档](https://docs.bigmodel.cn/cn/coding-plan/mcp/vision-mcp-server) | [English Document](https://docs.z.ai/devpack/mcp/vision-mcp-server)
-
-
-A Model Context Protocol (MCP) server that provides AI capabilities powered by Z.AI.
-
-## Environment Variables
-
-- `Z_AI_MODE` - The platform to use for AI services (default: `ZHIPU`)  ZHIPU or ZAI
-- `Z_AI_API_KEY` - Your API key
- 
-For ZAI: Use ZAI platform https://z.ai/model-api
-
-For ZHIPU: Use Zhipu AI platform https://bigmodel.cn
-
-## Usage
-
-### Use this MCP Server in Claude Code
-
-It is recommended to use the version Node.js 18+, Claude Code 2.0.14+.
-
-**Note**: You need to configure `Z_AI_MODE` as `ZAI` or `ZHIPU` depending on the platform.
-
-For ZAI Use,
-
-```shell
-claude mcp add zai-mcp-server --env Z_AI_API_KEY=your_api_key Z_AI_MODE=ZAI -- npx -y "@z_ai/mcp-server"
-```
-
-For ZHIPU Use,
-
-```shell
-claude mcp add zai-mcp-server --env Z_AI_API_KEY=your_api_key Z_AI_MODE=ZHIPU -- npx -y "@z_ai/mcp-server"
-```
-
-### Use this MCP Server in Others MCP Client
-
-For Z.ai Use, refer the [Vision MCP Doc](https://docs.z.ai/devpack/mcp/vision-mcp-server)
-
-For ZhiPU Use, refer the [Vision MCP Doc](https://docs.bigmodel.cn/cn/coding-plan/mcp/vision-mcp-server)

package/mcp-server/README.zh-CN.md

@@ -1,42 +0,0 @@
-# ZAI MCP 服务器
-
-[中文文档](https://docs.bigmodel.cn/cn/coding-plan/mcp/vision-mcp-server) | [English Document](https://docs.z.ai/devpack/mcp/vision-mcp-server)
-
-一个基于 Z.AI 提供 AI 能力的模型上下文协议 (MCP) 服务器。
-
-## MCP 工具
-
-该服务器实现了模型上下文协议，可与任何兼容 MCP 的客户端一起使用。服务器提供以下工具：
-
-- `image_analysis` - 分析图像并提供详细描述
-- `video_analysis` - 分析视频并提供详细描述
-
-## 环境变量
-
-- `Z_AI_MODE` - 用于 AI 服务的平台（默认：`ZHIPU`）ZHIPU 或 ZAI
-- `Z_AI_API_KEY` - 您的 API 密钥
-
-对于 `Z_AI_MODE=ZHIPU`：使用智谱 AI 平台 https://bigmodel.cn
-
-对于 `Z_AI_MODE=ZAI`：使用 ZAI 平台 https://z.ai/model-api 
-
-
-## 安装
-
-### 从 npm 安装
-
-```bash
-npm i @z_ai/mcp-server
-```
-
-## 使用方法
-
-### 在 Claude Code 中使用此 MCP 服务器
-
-```shell
-claude mcp add zai-mcp-server --env Z_AI_API_KEY=your_api_key  -- npx -y "@z_ai/mcp-server"
-```
-
-### 在其他客户端中使用此 MCP 服务器
-
-参考文档 [视觉理解 MCP](https://docs.bigmodel.cn/cn/coding-plan/mcp/vision-mcp-server)

package/mcp-server/build/core/api-common.js

@@ -1,122 +0,0 @@
-/**
- * Create multimodal message content
- * @param content Content array including images, text, etc.
- * @param prompt Text prompt
- * @returns Formatted message array
- */
-export function createMultiModalMessage(content, prompt) {
-    return [{
-            role: 'user',
-            content: [...content, { type: 'text', text: prompt }]
-        }];
-}
-/**
- * Create text message
- * @param prompt Text content
- * @returns Formatted message array
- */
-export function createTextMessage(prompt) {
-    return [{
-            role: 'user',
-            content: [{ type: 'text', text: prompt }]
-        }];
-}
-/**
- * Create image message content
- * @param imageUrl Image URL or base64 data
- * @returns Image content object
- */
-export function createImageContent(imageUrl) {
-    return {
-        type: 'image_url',
-        image_url: { url: imageUrl }
-    };
-}
-/**
- * Create video message content
- * @param videoUrl Video URL or base64 data
- * @returns Video content object
- */
-export function createVideoContent(videoUrl) {
-    return {
-        type: 'video_url', // Most AI models treat video as image_url type
-        video_url: { url: videoUrl }
-    };
-}
-/**
- * Create error response
- * @param message Error message
- * @param error Optional error object
- * @returns Standardized error response
- */
-export function createErrorResponse(message, error) {
-    return {
-        success: false,
-        error: message,
-        timestamp: Date.now(),
-        ...(error && { context: { stack: error.stack, name: error.name } })
-    };
-}
-/**
- * Create success response
- * @param data Response data
- * @returns Standardized success response
- */
-export function createSuccessResponse(data) {
-    return {
-        success: true,
-        data,
-        timestamp: Date.now()
-    };
-}
-/**
- * Format response content to MCP format
- * @param response API response
- * @returns MCP tool response format
- */
-export function formatMcpResponse(response) {
-    if (response.success) {
-        const text = typeof response.data === 'string'
-            ? response.data
-            : JSON.stringify(response.data, null, 2);
-        return {
-            content: [{ type: 'text', text }]
-        };
-    }
-    else {
-        return {
-            content: [{
-                    type: 'text',
-                    text: `Error: ${response.error}`
-                }],
-            isError: true
-        };
-    }
-}
-/**
- * Create async function with retry mechanism
- * @param fn Async function to execute
- * @param maxRetries Maximum retry attempts
- * @param delay Retry delay in milliseconds
- * @returns Wrapped function
- */
-export function withRetry(fn, maxRetries = 3, delay = 1000) {
-    return async (...args) => {
-        let lastError;
-        for (let attempt = 0; attempt <= maxRetries; attempt++) {
-            try {
-                return await fn(...args);
-            }
-            catch (error) {
-                lastError = error instanceof Error ? error : new Error(String(error));
-                if (attempt === maxRetries) {
-                    throw lastError;
-                }
-                // Exponential backoff
-                const waitTime = delay * Math.pow(2, attempt);
-                await new Promise(resolve => setTimeout(resolve, waitTime));
-            }
-        }
-        throw lastError;
-    };
-}

package/mcp-server/build/core/chat-service.js

@@ -1,80 +0,0 @@
-import { ApiError } from '../types/index.js';
-import { configurationService } from './environment.js';
-import { EnvironmentService } from './environment.js';
-/**
- * ZAI API service implementation
- */
-export class ChatService {
-    environmentService;
-    constructor(environmentService = EnvironmentService.getInstance()) {
-        this.environmentService = environmentService;
-    }
-    /**
-     * ZAI chat completions API for vision analysis
-     */
-    async visionCompletions(messages) {
-        const visionConfig = configurationService.getVisionConfig();
-        const requestBody = {
-            model: visionConfig.model,
-            messages,
-            thinking: { type: 'enabled' },
-            stream: false,
-            temperature: visionConfig.temperature,
-            top_p: visionConfig.topP,
-            max_tokens: visionConfig.maxTokens
-        };
-        console.info('Request ZAI chat completions API for vision analysis', { model: visionConfig.model, messageCount: messages.length });
-        try {
-            const response = await this.chatCompletions(visionConfig.url, requestBody);
-            const result = response.choices?.[0]?.message?.content;
-            if (!result) {
-                throw new ApiError('Invalid API response: missing content');
-            }
-            console.info('Request chat completions API for vision analysis successful');
-            return result;
-        }
-        catch (error) {
-            console.error('Request chat completions API for vision analysis failed', { error: error instanceof Error ? error.message : String(error) });
-            throw error instanceof ApiError ? error : new ApiError(`API call failed: ${error}`);
-        }
-    }
-    /**
-     * Make HTTP request to ZAI API with proper headers and error handling
-     */
-    async chatCompletions(url, body) {
-        const apiConfig = configurationService.getVisionConfig();
-        const apiKey = this.environmentService.getApiKey();
-        const controller = new AbortController();
-        const timeoutId = setTimeout(() => controller.abort(), apiConfig.timeout);
-        try {
-            const response = await fetch(url, {
-                method: 'POST',
-                headers: {
-                    'Authorization': `Bearer ${apiKey}`,
-                    'Content-Type': 'application/json',
-                    'X-Title': '4.5V MCP Local',
-                    'Accept-Language': 'en-US,en'
-                },
-                body: JSON.stringify(body),
-                signal: controller.signal
-            });
-            clearTimeout(timeoutId);
-            if (!response.ok) {
-                const errorText = await response.text();
-                throw new ApiError(`HTTP ${response.status}: ${errorText}`);
-            }
-            return await response.json();
-        }
-        catch (error) {
-            clearTimeout(timeoutId);
-            if (error instanceof ApiError) {
-                throw error;
-            }
-            throw new ApiError(`Network error: ${error}`);
-        }
-    }
-}
-/**
- * ZAI API chat completions service instance
- */
-export const chatService = new ChatService();