screenpipe-mcp 0.13.0 → 0.14.1

package/dist/http-server.js

@@ -86,89 +86,86 @@ async function fetchAPI(endpoint, options = {}) {
         },
     });
 }
-// Create MCP server
-const server = new index_js_1.Server({
-    name: "screenpipe-http",
-    version: "0.8.2",
-}, {
-    capabilities: {
-        tools: {},
-    },
-});
-// List tools handler
-server.setRequestHandler(types_js_1.ListToolsRequestSchema, async () => {
-    return { tools: TOOLS };
-});
-// Call tool handler
-server.setRequestHandler(types_js_1.CallToolRequestSchema, async (request) => {
-    const { name, arguments: args } = request.params;
-    if (!args) {
-        throw new Error("Missing arguments");
-    }
-    if (name === "search_content") {
-        const params = new URLSearchParams();
-        for (const [key, value] of Object.entries(args)) {
-            if (value !== null && value !== undefined) {
-                params.append(key, String(value));
-            }
-        }
-        const response = await fetchAPI(`/search?${params.toString()}`);
-        if (!response.ok) {
-            throw new Error(`HTTP error: ${response.status}`);
-        }
-        const data = await response.json();
-        const results = data.data || [];
-        const pagination = data.pagination || {};
-        if (results.length === 0) {
-            return {
-                content: [
-                    {
-                        type: "text",
-                        text: "No results found. Try: broader search terms, different content_type, or wider time range.",
-                    },
-                ],
-            };
-        }
-        const formattedResults = [];
-        for (const result of results) {
-            const content = result.content;
-            if (!content)
-                continue;
-            if (result.type === "OCR") {
-                formattedResults.push(`[OCR] ${content.app_name || "?"} | ${content.window_name || "?"}\n` +
-                    `${content.timestamp || ""}\n` +
-                    `${content.text || ""}`);
-            }
-            else if (result.type === "Audio") {
-                formattedResults.push(`[Audio] ${content.device_name || "?"}\n` +
-                    `${content.timestamp || ""}\n` +
-                    `${content.transcription || ""}`);
-            }
-            else if (result.type === "UI" || result.type === "Accessibility") {
-                formattedResults.push(`[Accessibility] ${content.app_name || "?"} | ${content.window_name || "?"}\n` +
-                    `${content.timestamp || ""}\n` +
-                    `${content.text || ""}`);
-            }
+// Tool handler for search_content
+async function handleSearchContent(args) {
+    const params = new URLSearchParams();
+    for (const [key, value] of Object.entries(args)) {
+        if (value !== null && value !== undefined) {
+            params.append(key, String(value));
         }
-        const header = `Results: ${results.length}/${pagination.total || "?"}` +
-            (pagination.total > results.length ? ` (use offset=${(pagination.offset || 0) + results.length} for more)` : "");
+    }
+    const response = await fetchAPI(`/search?${params.toString()}`);
+    if (!response.ok) {
+        throw new Error(`HTTP error: ${response.status}`);
+    }
+    const data = await response.json();
+    const results = data.data || [];
+    const pagination = data.pagination || {};
+    if (results.length === 0) {
         return {
             content: [
                 {
                     type: "text",
-                    text: header + "\n\n" + formattedResults.join("\n---\n"),
+                    text: "No results found. Try: broader search terms, different content_type, or wider time range.",
                 },
             ],
         };
     }
-    throw new Error(`Unknown tool: ${name}`);
-});
-// Create HTTP server with MCP transport
-const transports = new Map();
+    const formattedResults = [];
+    for (const result of results) {
+        const content = result.content;
+        if (!content)
+            continue;
+        if (result.type === "OCR") {
+            formattedResults.push(`[OCR] ${content.app_name || "?"} | ${content.window_name || "?"}\n` +
+                `${content.timestamp || ""}\n` +
+                `${content.text || ""}`);
+        }
+        else if (result.type === "Audio") {
+            formattedResults.push(`[Audio] ${content.device_name || "?"}\n` +
+                `${content.timestamp || ""}\n` +
+                `${content.transcription || ""}`);
+        }
+        else if (result.type === "UI" || result.type === "Accessibility") {
+            formattedResults.push(`[Accessibility] ${content.app_name || "?"} | ${content.window_name || "?"}\n` +
+                `${content.timestamp || ""}\n` +
+                `${content.text || ""}`);
+        }
+    }
+    const header = `Results: ${results.length}/${pagination.total || "?"}` +
+        (pagination.total > results.length ? ` (use offset=${(pagination.offset || 0) + results.length} for more)` : "");
+    return {
+        content: [
+            {
+                type: "text",
+                text: header + "\n\n" + formattedResults.join("\n---\n"),
+            },
+        ],
+    };
+}
+// Create a fresh MCP Server instance with handlers registered.
+// Each HTTP session gets its own Server — the MCP SDK requires a 1:1
+// mapping between Server and transport (reusing a Server across
+// transports throws "Already connected to a transport").
+function createMcpServer() {
+    const s = new index_js_1.Server({ name: "screenpipe-http", version: "0.14.0" }, { capabilities: { tools: {} } });
+    s.setRequestHandler(types_js_1.ListToolsRequestSchema, async () => ({ tools: TOOLS }));
+    s.setRequestHandler(types_js_1.CallToolRequestSchema, async (request) => {
+        const { name, arguments: args } = request.params;
+        if (!args)
+            throw new Error("Missing arguments");
+        if (name === "search_content")
+            return handleSearchContent(args);
+        throw new Error(`Unknown tool: ${name}`);
+    });
+    return s;
+}
+// Per-session state: each session gets its own Server + transport pair.
+const sessions = new Map();
 const httpServer = (0, http_1.createServer)(async (req, res) => {
     // CORS headers
     res.setHeader("Access-Control-Allow-Origin", "*");
-    res.setHeader("Access-Control-Allow-Methods", "GET, POST, OPTIONS");
+    res.setHeader("Access-Control-Allow-Methods", "GET, POST, DELETE, OPTIONS");
     res.setHeader("Access-Control-Allow-Headers", "Content-Type, Authorization, mcp-session-id");
     if (req.method === "OPTIONS") {
         res.writeHead(204);
@@ -178,23 +175,25 @@ const httpServer = (0, http_1.createServer)(async (req, res) => {
     // Health check
     if (req.url === "/health") {
         res.writeHead(200, { "Content-Type": "application/json" });
-        res.end(JSON.stringify({ status: "ok" }));
+        res.end(JSON.stringify({ status: "ok", sessions: sessions.size }));
         return;
     }
     // MCP endpoint
     if (req.url === "/mcp" || req.url?.startsWith("/mcp?")) {
         const sessionId = req.headers["mcp-session-id"];
-        let transport = sessionId ? transports.get(sessionId) : undefined;
-        if (!transport) {
-            transport = new streamableHttp_js_1.StreamableHTTPServerTransport({
+        let session = sessionId ? sessions.get(sessionId) : undefined;
+        if (!session) {
+            const server = createMcpServer();
+            const transport = new streamableHttp_js_1.StreamableHTTPServerTransport({
                 sessionIdGenerator: () => crypto.randomUUID(),
             });
             await server.connect(transport);
             if (transport.sessionId) {
-                transports.set(transport.sessionId, transport);
+                sessions.set(transport.sessionId, { server, transport });
             }
+            session = { server, transport };
         }
-        await transport.handleRequest(req, res);
+        await session.transport.handleRequest(req, res);
         return;
     }
     res.writeHead(404, { "Content-Type": "application/json" });

package/dist/index.js

@@ -136,9 +136,10 @@ const TOOLS = [
     },
     {
         name: "activity-summary",
-        description: "Lightweight activity overview (~200-500 tokens): app usage with active minutes, audio speakers, recent texts. " +
+        description: "Rich activity overview: app usage, window/tab titles with URLs and time spent, key text per context, audio transcriptions. " +
             "USE THIS FIRST for broad questions: 'what was I doing?', 'how long on X?', 'which apps?'. " +
-            "Only escalate to search-content if you need specific text content.",
+            "The 'windows' field shows exactly what the user worked on (e.g. 'Debug crash issue — 20 min', 'Stripe pricing page — 5 min'). " +
+            "Usually sufficient without further searches.",
         annotations: { title: "Activity Summary", readOnlyHint: true, openWorldHint: false, idempotentHint: true },
         inputSchema: {
             type: "object",
@@ -661,8 +662,19 @@ server.setRequestHandler(types_js_1.CallToolRequestSchema, async (request) => {
                         : "";
                     return `  ${a.name}: ${a.minutes} min (${a.frame_count} frames${timeSpan})`;
                 });
+                // Window/tab activity — what pages/documents were open
+                const windowLines = (data.windows || []).map((w) => {
+                    const url = w.browser_url ? ` (${w.browser_url})` : "";
+                    return `  [${w.app_name}] ${w.window_name}${url} — ${w.minutes} min`;
+                });
                 const speakerLines = (data.audio_summary?.speakers || []).map((s) => `  ${s.name}: ${s.segment_count} segments`);
-                const textLines = (data.recent_texts || []).map((t) => `  [${t.app_name}] ${t.text}`);
+                // Actual audio transcriptions (not just counts)
+                const transcriptLines = (data.audio_summary?.top_transcriptions || []).map((t) => `  [${t.speaker}, ${t.timestamp.slice(11, 19)}] ${t.transcription}`);
+                // Key text content sampled across the time range
+                const textLines = (data.key_texts || data.recent_texts || []).map((t) => {
+                    const win = t.window_name ? ` | ${t.window_name}` : "";
+                    return `  [${t.app_name}${win}, ${t.timestamp.slice(11, 19)}] ${t.text}`;
+                });
                 const summary = [
                     `Activity Summary (${data.time_range?.start} → ${data.time_range?.end})`,
                     `Total frames: ${data.total_frames}`,
@@ -670,11 +682,15 @@ server.setRequestHandler(types_js_1.CallToolRequestSchema, async (request) => {
                     "Apps:",
                     ...(appsLines.length ? appsLines : ["  (none)"]),
                     "",
+                    "Windows & Tabs:",
+                    ...(windowLines.length ? windowLines.slice(0, 20) : ["  (none)"]),
+                    "",
                     `Audio: ${data.audio_summary?.segment_count || 0} segments`,
                     ...(speakerLines.length ? speakerLines : []),
+                    ...(transcriptLines.length ? ["", "Audio transcriptions:", ...transcriptLines.slice(0, 15)] : []),
                     "",
-                    "Recent texts:",
-                    ...(textLines.length ? textLines.slice(0, 10) : ["  (none)"]),
+                    "Key content (sampled across time range):",
+                    ...(textLines.length ? textLines.slice(0, 20) : ["  (none)"]),
                 ].join("\n");
                 return { content: [{ type: "text", text: summary }] };
             }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "screenpipe-mcp",
-  "version": "0.13.0",
+  "version": "0.14.1",
   "description": "MCP server for screenpipe - search your screen recordings and audio transcriptions",
   "main": "dist/index.js",
   "bin": {

package/src/http-server.ts

@@ -96,109 +96,103 @@ async function fetchAPI(endpoint: string, options: RequestInit = {}): Promise<Re
   });
 }
 
-// Create MCP server
-const server = new Server(
-  {
-    name: "screenpipe-http",
-    version: "0.8.2",
-  },
-  {
-    capabilities: {
-      tools: {},
-    },
+// Tool handler for search_content
+async function handleSearchContent(args: Record<string, unknown>) {
+  const params = new URLSearchParams();
+  for (const [key, value] of Object.entries(args)) {
+    if (value !== null && value !== undefined) {
+      params.append(key, String(value));
+    }
   }
-);
-
-// List tools handler
-server.setRequestHandler(ListToolsRequestSchema, async () => {
-  return { tools: TOOLS };
-});
-
-// Call tool handler
-server.setRequestHandler(CallToolRequestSchema, async (request) => {
-  const { name, arguments: args } = request.params;
 
-  if (!args) {
-    throw new Error("Missing arguments");
+  const response = await fetchAPI(`/search?${params.toString()}`);
+  if (!response.ok) {
+    throw new Error(`HTTP error: ${response.status}`);
   }
 
-  if (name === "search_content") {
-    const params = new URLSearchParams();
-    for (const [key, value] of Object.entries(args)) {
-      if (value !== null && value !== undefined) {
-        params.append(key, String(value));
-      }
-    }
-
-    const response = await fetchAPI(`/search?${params.toString()}`);
-    if (!response.ok) {
-      throw new Error(`HTTP error: ${response.status}`);
-    }
-
-    const data = await response.json();
-    const results = data.data || [];
-    const pagination = data.pagination || {};
-
-    if (results.length === 0) {
-      return {
-        content: [
-          {
-            type: "text",
-            text: "No results found. Try: broader search terms, different content_type, or wider time range.",
-          },
-        ],
-      };
-    }
-
-    const formattedResults: string[] = [];
-    for (const result of results) {
-      const content = result.content;
-      if (!content) continue;
-
-      if (result.type === "OCR") {
-        formattedResults.push(
-          `[OCR] ${content.app_name || "?"} | ${content.window_name || "?"}\n` +
-          `${content.timestamp || ""}\n` +
-          `${content.text || ""}`
-        );
-      } else if (result.type === "Audio") {
-        formattedResults.push(
-          `[Audio] ${content.device_name || "?"}\n` +
-          `${content.timestamp || ""}\n` +
-          `${content.transcription || ""}`
-        );
-      } else if (result.type === "UI" || result.type === "Accessibility") {
-        formattedResults.push(
-          `[Accessibility] ${content.app_name || "?"} | ${content.window_name || "?"}\n` +
-          `${content.timestamp || ""}\n` +
-          `${content.text || ""}`
-        );
-      }
-    }
-
-    const header = `Results: ${results.length}/${pagination.total || "?"}` +
-      (pagination.total > results.length ? ` (use offset=${(pagination.offset || 0) + results.length} for more)` : "");
+  const data = await response.json();
+  const results = data.data || [];
+  const pagination = data.pagination || {};
 
+  if (results.length === 0) {
     return {
       content: [
         {
           type: "text",
-          text: header + "\n\n" + formattedResults.join("\n---\n"),
+          text: "No results found. Try: broader search terms, different content_type, or wider time range.",
         },
       ],
     };
   }
 
-  throw new Error(`Unknown tool: ${name}`);
-});
+  const formattedResults: string[] = [];
+  for (const result of results) {
+    const content = result.content;
+    if (!content) continue;
+
+    if (result.type === "OCR") {
+      formattedResults.push(
+        `[OCR] ${content.app_name || "?"} | ${content.window_name || "?"}\n` +
+        `${content.timestamp || ""}\n` +
+        `${content.text || ""}`
+      );
+    } else if (result.type === "Audio") {
+      formattedResults.push(
+        `[Audio] ${content.device_name || "?"}\n` +
+        `${content.timestamp || ""}\n` +
+        `${content.transcription || ""}`
+      );
+    } else if (result.type === "UI" || result.type === "Accessibility") {
+      formattedResults.push(
+        `[Accessibility] ${content.app_name || "?"} | ${content.window_name || "?"}\n` +
+        `${content.timestamp || ""}\n` +
+        `${content.text || ""}`
+      );
+    }
+  }
+
+  const header = `Results: ${results.length}/${pagination.total || "?"}` +
+    (pagination.total > results.length ? ` (use offset=${(pagination.offset || 0) + results.length} for more)` : "");
+
+  return {
+    content: [
+      {
+        type: "text",
+        text: header + "\n\n" + formattedResults.join("\n---\n"),
+      },
+    ],
+  };
+}
+
+// Create a fresh MCP Server instance with handlers registered.
+// Each HTTP session gets its own Server — the MCP SDK requires a 1:1
+// mapping between Server and transport (reusing a Server across
+// transports throws "Already connected to a transport").
+function createMcpServer(): Server {
+  const s = new Server(
+    { name: "screenpipe-http", version: "0.14.0" },
+    { capabilities: { tools: {} } }
+  );
+
+  s.setRequestHandler(ListToolsRequestSchema, async () => ({ tools: TOOLS }));
+
+  s.setRequestHandler(CallToolRequestSchema, async (request) => {
+    const { name, arguments: args } = request.params;
+    if (!args) throw new Error("Missing arguments");
+    if (name === "search_content") return handleSearchContent(args);
+    throw new Error(`Unknown tool: ${name}`);
+  });
+
+  return s;
+}
 
-// Create HTTP server with MCP transport
-const transports = new Map<string, StreamableHTTPServerTransport>();
+// Per-session state: each session gets its own Server + transport pair.
+const sessions = new Map<string, { server: Server; transport: StreamableHTTPServerTransport }>();
 
 const httpServer = createServer(async (req, res) => {
   // CORS headers
   res.setHeader("Access-Control-Allow-Origin", "*");
-  res.setHeader("Access-Control-Allow-Methods", "GET, POST, OPTIONS");
+  res.setHeader("Access-Control-Allow-Methods", "GET, POST, DELETE, OPTIONS");
   res.setHeader("Access-Control-Allow-Headers", "Content-Type, Authorization, mcp-session-id");
 
   if (req.method === "OPTIONS") {
@@ -210,7 +204,7 @@ const httpServer = createServer(async (req, res) => {
   // Health check
   if (req.url === "/health") {
     res.writeHead(200, { "Content-Type": "application/json" });
-    res.end(JSON.stringify({ status: "ok" }));
+    res.end(JSON.stringify({ status: "ok", sessions: sessions.size }));
     return;
   }
 
@@ -218,21 +212,23 @@ const httpServer = createServer(async (req, res) => {
   if (req.url === "/mcp" || req.url?.startsWith("/mcp?")) {
     const sessionId = req.headers["mcp-session-id"] as string | undefined;
 
-    let transport = sessionId ? transports.get(sessionId) : undefined;
+    let session = sessionId ? sessions.get(sessionId) : undefined;
 
-    if (!transport) {
-      transport = new StreamableHTTPServerTransport({
+    if (!session) {
+      const server = createMcpServer();
+      const transport = new StreamableHTTPServerTransport({
         sessionIdGenerator: () => crypto.randomUUID(),
       });
 
       await server.connect(transport);
 
       if (transport.sessionId) {
-        transports.set(transport.sessionId, transport);
+        sessions.set(transport.sessionId, { server, transport });
       }
+      session = { server, transport };
     }
 
-    await transport.handleRequest(req, res);
+    await session.transport.handleRequest(req, res);
     return;
   }
 

package/src/index.ts

@@ -119,9 +119,10 @@ const TOOLS: Tool[] = [
   {
     name: "activity-summary",
     description:
-      "Lightweight activity overview (~200-500 tokens): app usage with active minutes, audio speakers, recent texts. " +
+      "Rich activity overview: app usage, window/tab titles with URLs and time spent, key text per context, audio transcriptions. " +
       "USE THIS FIRST for broad questions: 'what was I doing?', 'how long on X?', 'which apps?'. " +
-      "Only escalate to search-content if you need specific text content.",
+      "The 'windows' field shows exactly what the user worked on (e.g. 'Debug crash issue — 20 min', 'Stripe pricing page — 5 min'). " +
+      "Usually sufficient without further searches.",
     annotations: { title: "Activity Summary", readOnlyHint: true, openWorldHint: false, idempotentHint: true },
     inputSchema: {
       type: "object",
@@ -709,14 +710,37 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
           }
         );
 
+        // Window/tab activity — what pages/documents were open
+        const windowLines = (data.windows || []).map(
+          (w: {
+            app_name: string;
+            window_name: string;
+            browser_url: string;
+            minutes: number;
+            frame_count: number;
+          }) => {
+            const url = w.browser_url ? ` (${w.browser_url})` : "";
+            return `  [${w.app_name}] ${w.window_name}${url} — ${w.minutes} min`;
+          }
+        );
+
         const speakerLines = (data.audio_summary?.speakers || []).map(
           (s: { name: string; segment_count: number }) =>
             `  ${s.name}: ${s.segment_count} segments`
         );
 
-        const textLines = (data.recent_texts || []).map(
-          (t: { text: string; app_name: string; timestamp: string }) =>
-            `  [${t.app_name}] ${t.text}`
+        // Actual audio transcriptions (not just counts)
+        const transcriptLines = (data.audio_summary?.top_transcriptions || []).map(
+          (t: { transcription: string; speaker: string; device: string; timestamp: string }) =>
+            `  [${t.speaker}, ${t.timestamp.slice(11, 19)}] ${t.transcription}`
+        );
+
+        // Key text content sampled across the time range
+        const textLines = (data.key_texts || data.recent_texts || []).map(
+          (t: { text: string; app_name: string; window_name?: string; timestamp: string }) => {
+            const win = t.window_name ? ` | ${t.window_name}` : "";
+            return `  [${t.app_name}${win}, ${t.timestamp.slice(11, 19)}] ${t.text}`;
+          }
         );
 
         const summary = [
@@ -726,11 +750,15 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
           "Apps:",
           ...(appsLines.length ? appsLines : ["  (none)"]),
           "",
+          "Windows & Tabs:",
+          ...(windowLines.length ? windowLines.slice(0, 20) : ["  (none)"]),
+          "",
           `Audio: ${data.audio_summary?.segment_count || 0} segments`,
           ...(speakerLines.length ? speakerLines : []),
+          ...(transcriptLines.length ? ["", "Audio transcriptions:", ...transcriptLines.slice(0, 15)] : []),
           "",
-          "Recent texts:",
-          ...(textLines.length ? textLines.slice(0, 10) : ["  (none)"]),
+          "Key content (sampled across time range):",
+          ...(textLines.length ? textLines.slice(0, 20) : ["  (none)"]),
         ].join("\n");
 
         return { content: [{ type: "text", text: summary }] };