notioncode 0.1.0 → 0.1.2

package/agent-runtime-server/server/projects.js

@@ -0,0 +1,2594 @@
+/**
+ * PROJECT DISCOVERY AND MANAGEMENT SYSTEM
+ * ========================================
+ * 
+ * This module manages project discovery for both Claude CLI and Cursor CLI sessions.
+ * 
+ * ## Architecture Overview
+ * 
+ * 1. **Claude Projects** (stored in ~/.claude/projects/)
+ *    - Each project is a directory named with the project path encoded (/ replaced with -)
+ *    - Contains .jsonl files with conversation history including 'cwd' field
+ *    - Project metadata stored in ~/.claude/project-config.json
+ * 
+ * 2. **Cursor Projects** (stored in ~/.cursor/chats/)
+ *    - Each project directory is named with MD5 hash of the absolute project path
+ *    - Example: /Users/john/myproject -> MD5 -> a1b2c3d4e5f6...
+ *    - Contains session directories with SQLite databases (store.db)
+ *    - Project path is NOT stored in the database - only in the MD5 hash
+ * 
+ * ## Project Discovery Strategy
+ * 
+ * 1. **Claude Projects Discovery**:
+ *    - Scan ~/.claude/projects/ directory for Claude project folders
+ *    - Extract actual project path from .jsonl files (cwd field)
+ *    - Fall back to decoded directory name if no sessions exist
+ * 
+ * 2. **Cursor Sessions Discovery**:
+ *    - For each KNOWN project (from Claude or manually added)
+ *    - Compute MD5 hash of the project's absolute path
+ *    - Check if ~/.cursor/chats/{md5_hash}/ directory exists
+ *    - Read session metadata from SQLite store.db files
+ * 
+ * 3. **Manual Project Addition**:
+ *    - Users can manually add project paths via UI
+ *    - Stored in ~/.claude/project-config.json with 'manuallyAdded' flag
+ *    - Allows discovering Cursor sessions for projects without Claude sessions
+ * 
+ * ## Critical Limitations
+ * 
+ * - **CANNOT discover Cursor-only projects**: From a quick check, there was no mention of
+ *   the cwd of each project. if someone has the time, you can try to reverse engineer it.
+ * 
+ * - **Project relocation breaks history**: If a project directory is moved or renamed,
+ *   the MD5 hash changes, making old Cursor sessions inaccessible unless the old
+ *   path is known and manually added.
+ * 
+ * ## Error Handling
+ * 
+ * - Missing ~/.claude directory is handled gracefully with automatic creation
+ * - ENOENT errors are caught and handled without crashing
+ * - Empty arrays returned when no projects/sessions exist
+ * 
+ * ## Caching Strategy
+ * 
+ * - Project directory extraction is cached to minimize file I/O
+ * - Cache is cleared when project configuration changes
+ * - Session data is fetched on-demand, not cached
+ */
+
+import { promises as fs } from 'fs';
+import fsSync from 'fs';
+import path from 'path';
+import readline from 'readline';
+import crypto from 'crypto';
+import sqlite3 from 'sqlite3';
+import { open } from 'sqlite';
+import os from 'os';
+import sessionManager from './sessionManager.js';
+import { applyCustomSessionNames } from './database/db.js';
+
+// Import TaskMaster detection functions
+async function detectTaskMasterFolder(projectPath) {
+  try {
+    const taskMasterPath = path.join(projectPath, '.taskmaster');
+
+    // Check if .taskmaster directory exists
+    try {
+      const stats = await fs.stat(taskMasterPath);
+      if (!stats.isDirectory()) {
+        return {
+          hasTaskmaster: false,
+          reason: '.taskmaster exists but is not a directory'
+        };
+      }
+    } catch (error) {
+      if (error.code === 'ENOENT') {
+        return {
+          hasTaskmaster: false,
+          reason: '.taskmaster directory not found'
+        };
+      }
+      throw error;
+    }
+
+    // Check for key TaskMaster files
+    const keyFiles = [
+      'tasks/tasks.json',
+      'config.json'
+    ];
+
+    const fileStatus = {};
+    let hasEssentialFiles = true;
+
+    for (const file of keyFiles) {
+      const filePath = path.join(taskMasterPath, file);
+      try {
+        await fs.access(filePath);
+        fileStatus[file] = true;
+      } catch (error) {
+        fileStatus[file] = false;
+        if (file === 'tasks/tasks.json') {
+          hasEssentialFiles = false;
+        }
+      }
+    }
+
+    // Parse tasks.json if it exists for metadata
+    let taskMetadata = null;
+    if (fileStatus['tasks/tasks.json']) {
+      try {
+        const tasksPath = path.join(taskMasterPath, 'tasks/tasks.json');
+        const tasksContent = await fs.readFile(tasksPath, 'utf8');
+        const tasksData = JSON.parse(tasksContent);
+
+        // Handle both tagged and legacy formats
+        let tasks = [];
+        if (tasksData.tasks) {
+          // Legacy format
+          tasks = tasksData.tasks;
+        } else {
+          // Tagged format - get tasks from all tags
+          Object.values(tasksData).forEach(tagData => {
+            if (tagData.tasks) {
+              tasks = tasks.concat(tagData.tasks);
+            }
+          });
+        }
+
+        // Calculate task statistics
+        const stats = tasks.reduce((acc, task) => {
+          acc.total++;
+          acc[task.status] = (acc[task.status] || 0) + 1;
+
+          // Count subtasks
+          if (task.subtasks) {
+            task.subtasks.forEach(subtask => {
+              acc.subtotalTasks++;
+              acc.subtasks = acc.subtasks || {};
+              acc.subtasks[subtask.status] = (acc.subtasks[subtask.status] || 0) + 1;
+            });
+          }
+
+          return acc;
+        }, {
+          total: 0,
+          subtotalTasks: 0,
+          pending: 0,
+          'in-progress': 0,
+          done: 0,
+          review: 0,
+          deferred: 0,
+          cancelled: 0,
+          subtasks: {}
+        });
+
+        taskMetadata = {
+          taskCount: stats.total,
+          subtaskCount: stats.subtotalTasks,
+          completed: stats.done || 0,
+          pending: stats.pending || 0,
+          inProgress: stats['in-progress'] || 0,
+          review: stats.review || 0,
+          completionPercentage: stats.total > 0 ? Math.round((stats.done / stats.total) * 100) : 0,
+          lastModified: (await fs.stat(tasksPath)).mtime.toISOString()
+        };
+      } catch (parseError) {
+        console.warn('Failed to parse tasks.json:', parseError.message);
+        taskMetadata = { error: 'Failed to parse tasks.json' };
+      }
+    }
+
+    return {
+      hasTaskmaster: true,
+      hasEssentialFiles,
+      files: fileStatus,
+      metadata: taskMetadata,
+      path: taskMasterPath
+    };
+
+  } catch (error) {
+    console.error('Error detecting TaskMaster folder:', error);
+    return {
+      hasTaskmaster: false,
+      reason: `Error checking directory: ${error.message}`
+    };
+  }
+}
+
+// Cache for extracted project directories
+const projectDirectoryCache = new Map();
+const staleProjectWarnings = new Set();
+
+async function pathExists(targetPath) {
+  if (!targetPath) return false;
+  try {
+    await fs.access(targetPath);
+    return true;
+  } catch {
+    return false;
+  }
+}
+
+// Clear cache when needed (called when project files change)
+function clearProjectDirectoryCache() {
+  projectDirectoryCache.clear();
+}
+
+function warnStaleProjectOnce(projectName, resolvedPath, action = 'Skipping stale project') {
+  const key = `${projectName}:${resolvedPath}:${action}`;
+  if (staleProjectWarnings.has(key)) {
+    return;
+  }
+  staleProjectWarnings.add(key);
+  if (process.env.DEBUG_PROJECT_DISCOVERY === 'true') {
+    console.warn(`[Projects] ${action} ${projectName}: resolved path missing (${resolvedPath})`);
+  }
+}
+
+// Load project configuration file
+async function loadProjectConfig() {
+  const configPath = path.join(os.homedir(), '.claude', 'project-config.json');
+  try {
+    const configData = await fs.readFile(configPath, 'utf8');
+    return JSON.parse(configData);
+  } catch (error) {
+    // Return empty config if file doesn't exist
+    return {};
+  }
+}
+
+// Save project configuration file
+async function saveProjectConfig(config) {
+  const claudeDir = path.join(os.homedir(), '.claude');
+  const configPath = path.join(claudeDir, 'project-config.json');
+
+  // Ensure the .claude directory exists
+  try {
+    await fs.mkdir(claudeDir, { recursive: true });
+  } catch (error) {
+    if (error.code !== 'EEXIST') {
+      throw error;
+    }
+  }
+
+  await fs.writeFile(configPath, JSON.stringify(config, null, 2), 'utf8');
+}
+
+// Generate better display name from path
+async function generateDisplayName(projectName, actualProjectDir = null) {
+  // Use actual project directory if provided, otherwise decode from project name
+  let projectPath = actualProjectDir || projectName.replace(/-/g, '/');
+
+  // Try to read package.json from the project path
+  try {
+    const packageJsonPath = path.join(projectPath, 'package.json');
+    const packageData = await fs.readFile(packageJsonPath, 'utf8');
+    const packageJson = JSON.parse(packageData);
+
+    // Return the name from package.json if it exists
+    if (packageJson.name) {
+      return packageJson.name;
+    }
+  } catch (error) {
+    // Fall back to path-based naming if package.json doesn't exist or can't be read
+  }
+
+  // If it starts with /, it's an absolute path
+  if (projectPath.startsWith('/')) {
+    const parts = projectPath.split('/').filter(Boolean);
+    // Return only the last folder name
+    return parts[parts.length - 1] || projectPath;
+  }
+
+  return projectPath;
+}
+
+// Extract the actual project directory from JSONL sessions (with caching)
+async function extractProjectDirectory(projectName) {
+  // Check cache first
+  if (projectDirectoryCache.has(projectName)) {
+    return projectDirectoryCache.get(projectName);
+  }
+
+  // Check project config for originalPath (manually added projects via UI or platform)
+  // This handles projects with dashes in their directory names correctly
+  const config = await loadProjectConfig();
+  if (config[projectName]?.originalPath) {
+    const originalPath = config[projectName].originalPath;
+    projectDirectoryCache.set(projectName, originalPath);
+    return originalPath;
+  }
+
+  const projectDir = path.join(os.homedir(), '.claude', 'projects', projectName);
+  const cwdCounts = new Map();
+  let latestTimestamp = 0;
+  let latestCwd = null;
+  let extractedPath;
+
+  try {
+    // Check if the project directory exists
+    await fs.access(projectDir);
+
+    const files = await fs.readdir(projectDir);
+    const jsonlFiles = files.filter(file => file.endsWith('.jsonl'));
+
+    if (jsonlFiles.length === 0) {
+      // Fall back to decoded project name if no sessions
+      extractedPath = projectName.replace(/-/g, '/');
+    } else {
+      // Process all JSONL files to collect cwd values
+      for (const file of jsonlFiles) {
+        const jsonlFile = path.join(projectDir, file);
+        const fileStream = fsSync.createReadStream(jsonlFile);
+        const rl = readline.createInterface({
+          input: fileStream,
+          crlfDelay: Infinity
+        });
+
+        for await (const line of rl) {
+          if (line.trim()) {
+            try {
+              const entry = JSON.parse(line);
+
+              if (entry.cwd) {
+                // Count occurrences of each cwd
+                cwdCounts.set(entry.cwd, (cwdCounts.get(entry.cwd) || 0) + 1);
+
+                // Track the most recent cwd
+                const timestamp = new Date(entry.timestamp || 0).getTime();
+                if (timestamp > latestTimestamp) {
+                  latestTimestamp = timestamp;
+                  latestCwd = entry.cwd;
+                }
+              }
+            } catch (parseError) {
+              // Skip malformed lines
+            }
+          }
+        }
+      }
+
+      // Determine the best cwd to use
+      if (cwdCounts.size === 0) {
+        // No cwd found, fall back to decoded project name
+        extractedPath = projectName.replace(/-/g, '/');
+      } else if (cwdCounts.size === 1) {
+        // Only one cwd, use it
+        extractedPath = Array.from(cwdCounts.keys())[0];
+      } else {
+        // Multiple cwd values - prefer the most recent one if it has reasonable usage
+        const mostRecentCount = cwdCounts.get(latestCwd) || 0;
+        const maxCount = Math.max(...cwdCounts.values());
+
+        // Use most recent if it has at least 25% of the max count
+        if (mostRecentCount >= maxCount * 0.25) {
+          extractedPath = latestCwd;
+        } else {
+          // Otherwise use the most frequently used cwd
+          for (const [cwd, count] of cwdCounts.entries()) {
+            if (count === maxCount) {
+              extractedPath = cwd;
+              break;
+            }
+          }
+        }
+
+        // Fallback (shouldn't reach here)
+        if (!extractedPath) {
+          extractedPath = latestCwd || projectName.replace(/-/g, '/');
+        }
+      }
+    }
+
+    // Cache the result
+    projectDirectoryCache.set(projectName, extractedPath);
+
+    return extractedPath;
+
+  } catch (error) {
+    // If the directory doesn't exist, just use the decoded project name
+    if (error.code === 'ENOENT') {
+      extractedPath = projectName.replace(/-/g, '/');
+    } else {
+      console.error(`Error extracting project directory for ${projectName}:`, error);
+      // Fall back to decoded project name for other errors
+      extractedPath = projectName.replace(/-/g, '/');
+    }
+
+    // Cache the fallback result too
+    projectDirectoryCache.set(projectName, extractedPath);
+
+    return extractedPath;
+  }
+}
+
+async function getProjects(progressCallback = null) {
+  const claudeDir = path.join(os.homedir(), '.claude', 'projects');
+  const config = await loadProjectConfig();
+  const projects = [];
+  const existingProjects = new Set();
+  const codexSessionsIndexRef = { sessionsByProject: null };
+  let totalProjects = 0;
+  let processedProjects = 0;
+  let directories = [];
+
+  try {
+    // Check if the .claude/projects directory exists
+    await fs.access(claudeDir);
+
+    // First, get existing Claude projects from the file system
+    const entries = await fs.readdir(claudeDir, { withFileTypes: true });
+    directories = entries.filter(e => e.isDirectory());
+
+    // Build set of existing project names for later
+    directories.forEach(e => existingProjects.add(e.name));
+
+    // Count manual projects not already in directories
+    const manualProjectsCount = Object.entries(config)
+      .filter(([name, cfg]) => cfg.manuallyAdded && !existingProjects.has(name))
+      .length;
+
+    totalProjects = directories.length + manualProjectsCount;
+
+    for (const entry of directories) {
+      processedProjects++;
+
+      // Emit progress
+      if (progressCallback) {
+        progressCallback({
+          phase: 'loading',
+          current: processedProjects,
+          total: totalProjects,
+          currentProject: entry.name
+        });
+      }
+
+      // Extract actual project directory from JSONL sessions
+      const actualProjectDir = await extractProjectDirectory(entry.name);
+      if (!(await pathExists(actualProjectDir))) {
+        warnStaleProjectOnce(entry.name, actualProjectDir);
+        continue;
+      }
+
+      // Get display name from config or generate one
+      const customName = config[entry.name]?.displayName;
+      const autoDisplayName = await generateDisplayName(entry.name, actualProjectDir);
+      const fullPath = actualProjectDir;
+
+      const project = {
+        name: entry.name,
+        path: actualProjectDir,
+        displayName: customName || autoDisplayName,
+        fullPath: fullPath,
+        isCustomName: !!customName,
+        sessions: [],
+        geminiSessions: [],
+        sessionMeta: {
+          hasMore: false,
+          total: 0
+        }
+      };
+
+      // Try to get sessions for this project (just first 5 for performance)
+      try {
+        const sessionResult = await getSessions(entry.name, 5, 0);
+        project.sessions = sessionResult.sessions || [];
+        project.sessionMeta = {
+          hasMore: sessionResult.hasMore,
+          total: sessionResult.total
+        };
+      } catch (e) {
+        console.warn(`Could not load sessions for project ${entry.name}:`, e.message);
+        project.sessionMeta = {
+          hasMore: false,
+          total: 0
+        };
+      }
+      applyCustomSessionNames(project.sessions, 'claude');
+
+      // Also fetch Cursor sessions for this project
+      try {
+        project.cursorSessions = await getCursorSessions(actualProjectDir);
+      } catch (e) {
+        console.warn(`Could not load Cursor sessions for project ${entry.name}:`, e.message);
+        project.cursorSessions = [];
+      }
+      applyCustomSessionNames(project.cursorSessions, 'cursor');
+
+      // Also fetch Codex sessions for this project
+      try {
+        project.codexSessions = await getCodexSessions(actualProjectDir, {
+          indexRef: codexSessionsIndexRef,
+        });
+      } catch (e) {
+        console.warn(`Could not load Codex sessions for project ${entry.name}:`, e.message);
+        project.codexSessions = [];
+      }
+      applyCustomSessionNames(project.codexSessions, 'codex');
+
+      // Also fetch Gemini sessions for this project (UI + CLI)
+      try {
+        const uiSessions = sessionManager.getProjectSessions(actualProjectDir) || [];
+        const cliSessions = await getGeminiCliSessions(actualProjectDir);
+        const uiIds = new Set(uiSessions.map(s => s.id));
+        const mergedGemini = [...uiSessions, ...cliSessions.filter(s => !uiIds.has(s.id))];
+        project.geminiSessions = mergedGemini;
+      } catch (e) {
+        console.warn(`Could not load Gemini sessions for project ${entry.name}:`, e.message);
+        project.geminiSessions = [];
+      }
+      applyCustomSessionNames(project.geminiSessions, 'gemini');
+
+      // Add TaskMaster detection
+      try {
+        const taskMasterResult = await detectTaskMasterFolder(actualProjectDir);
+        project.taskmaster = {
+          hasTaskmaster: taskMasterResult.hasTaskmaster,
+          hasEssentialFiles: taskMasterResult.hasEssentialFiles,
+          metadata: taskMasterResult.metadata,
+          status: taskMasterResult.hasTaskmaster && taskMasterResult.hasEssentialFiles ? 'configured' : 'not-configured'
+        };
+      } catch (e) {
+        console.warn(`Could not detect TaskMaster for project ${entry.name}:`, e.message);
+        project.taskmaster = {
+          hasTaskmaster: false,
+          hasEssentialFiles: false,
+          metadata: null,
+          status: 'error'
+        };
+      }
+
+      projects.push(project);
+    }
+  } catch (error) {
+    // If the directory doesn't exist (ENOENT), that's okay - just continue with empty projects
+    if (error.code !== 'ENOENT') {
+      console.error('Error reading projects directory:', error);
+    }
+    // Calculate total for manual projects only (no directories exist)
+    totalProjects = Object.entries(config)
+      .filter(([name, cfg]) => cfg.manuallyAdded)
+      .length;
+  }
+
+  // Add manually configured projects that don't exist as folders yet
+  for (const [projectName, projectConfig] of Object.entries(config)) {
+    if (!existingProjects.has(projectName) && projectConfig.manuallyAdded) {
+      processedProjects++;
+
+      // Emit progress for manual projects
+      if (progressCallback) {
+        progressCallback({
+          phase: 'loading',
+          current: processedProjects,
+          total: totalProjects,
+          currentProject: projectName
+        });
+      }
+
+      // Use the original path if available, otherwise extract from potential sessions
+      let actualProjectDir = projectConfig.originalPath;
+
+      if (!actualProjectDir) {
+        try {
+          actualProjectDir = await extractProjectDirectory(projectName);
+        } catch (error) {
+          // Fall back to decoded project name
+          actualProjectDir = projectName.replace(/-/g, '/');
+        }
+      }
+
+      if (!(await pathExists(actualProjectDir))) {
+        warnStaleProjectOnce(projectName, actualProjectDir, 'Removing stale manual project');
+        delete config[projectName];
+        await saveProjectConfig(config);
+        continue;
+      }
+
+      const project = {
+        name: projectName,
+        path: actualProjectDir,
+        displayName: projectConfig.displayName || await generateDisplayName(projectName, actualProjectDir),
+        fullPath: actualProjectDir,
+        isCustomName: !!projectConfig.displayName,
+        isManuallyAdded: true,
+        sessions: [],
+        geminiSessions: [],
+        sessionMeta: {
+          hasMore: false,
+          total: 0
+        },
+        cursorSessions: [],
+        codexSessions: []
+      };
+
+      // Try to fetch Cursor sessions for manual projects too
+      try {
+        project.cursorSessions = await getCursorSessions(actualProjectDir);
+      } catch (e) {
+        console.warn(`Could not load Cursor sessions for manual project ${projectName}:`, e.message);
+      }
+      applyCustomSessionNames(project.cursorSessions, 'cursor');
+
+      // Try to fetch Codex sessions for manual projects too
+      try {
+        project.codexSessions = await getCodexSessions(actualProjectDir, {
+          indexRef: codexSessionsIndexRef,
+        });
+      } catch (e) {
+        console.warn(`Could not load Codex sessions for manual project ${projectName}:`, e.message);
+      }
+      applyCustomSessionNames(project.codexSessions, 'codex');
+
+      // Try to fetch Gemini sessions for manual projects too (UI + CLI)
+      try {
+        const uiSessions = sessionManager.getProjectSessions(actualProjectDir) || [];
+        const cliSessions = await getGeminiCliSessions(actualProjectDir);
+        const uiIds = new Set(uiSessions.map(s => s.id));
+        project.geminiSessions = [...uiSessions, ...cliSessions.filter(s => !uiIds.has(s.id))];
+      } catch (e) {
+        console.warn(`Could not load Gemini sessions for manual project ${projectName}:`, e.message);
+      }
+      applyCustomSessionNames(project.geminiSessions, 'gemini');
+
+      // Add TaskMaster detection for manual projects
+      try {
+        const taskMasterResult = await detectTaskMasterFolder(actualProjectDir);
+
+        // Determine TaskMaster status
+        let taskMasterStatus = 'not-configured';
+        if (taskMasterResult.hasTaskmaster && taskMasterResult.hasEssentialFiles) {
+          taskMasterStatus = 'taskmaster-only'; // We don't check MCP for manual projects in bulk
+        }
+
+        project.taskmaster = {
+          status: taskMasterStatus,
+          hasTaskmaster: taskMasterResult.hasTaskmaster,
+          hasEssentialFiles: taskMasterResult.hasEssentialFiles,
+          metadata: taskMasterResult.metadata
+        };
+      } catch (error) {
+        console.warn(`TaskMaster detection failed for manual project ${projectName}:`, error.message);
+        project.taskmaster = {
+          status: 'error',
+          hasTaskmaster: false,
+          hasEssentialFiles: false,
+          error: error.message
+        };
+      }
+
+      projects.push(project);
+    }
+  }
+
+  // Emit completion after all projects (including manual) are processed
+  if (progressCallback) {
+    progressCallback({
+      phase: 'complete',
+      current: totalProjects,
+      total: totalProjects
+    });
+  }
+
+  return projects;
+}
+
+async function getSessions(projectName, limit = 5, offset = 0) {
+  const projectDir = path.join(os.homedir(), '.claude', 'projects', projectName);
+
+  try {
+    const files = await fs.readdir(projectDir);
+    // agent-*.jsonl files contain session start data at this point. This needs to be revisited
+    // periodically to make sure only accurate data is there and no new functionality is added there
+    const jsonlFiles = files.filter(file => file.endsWith('.jsonl') && !file.startsWith('agent-'));
+
+    if (jsonlFiles.length === 0) {
+      return { sessions: [], hasMore: false, total: 0 };
+    }
+
+    // Sort files by modification time (newest first)
+    const filesWithStats = await Promise.all(
+      jsonlFiles.map(async (file) => {
+        const filePath = path.join(projectDir, file);
+        const stats = await fs.stat(filePath);
+        return { file, mtime: stats.mtime };
+      })
+    );
+    filesWithStats.sort((a, b) => b.mtime - a.mtime);
+
+    const allSessions = new Map();
+    const allEntries = [];
+    const uuidToSessionMap = new Map();
+
+    // Collect all sessions and entries from all files
+    for (const { file } of filesWithStats) {
+      const jsonlFile = path.join(projectDir, file);
+      const result = await parseJsonlSessions(jsonlFile);
+
+      result.sessions.forEach(session => {
+        if (!allSessions.has(session.id)) {
+          allSessions.set(session.id, session);
+        }
+      });
+
+      allEntries.push(...result.entries);
+
+      // Early exit optimization for large projects
+      if (allSessions.size >= (limit + offset) * 2 && allEntries.length >= Math.min(3, filesWithStats.length)) {
+        break;
+      }
+    }
+
+    // Build UUID-to-session mapping for timeline detection
+    allEntries.forEach(entry => {
+      if (entry.uuid && entry.sessionId) {
+        uuidToSessionMap.set(entry.uuid, entry.sessionId);
+      }
+    });
+
+    // Group sessions by first user message ID
+    const sessionGroups = new Map(); // firstUserMsgId -> { latestSession, allSessions[] }
+    const sessionToFirstUserMsgId = new Map(); // sessionId -> firstUserMsgId
+
+    // Find the first user message for each session
+    allEntries.forEach(entry => {
+      if (entry.sessionId && entry.type === 'user' && entry.parentUuid === null && entry.uuid) {
+        // This is a first user message in a session (parentUuid is null)
+        const firstUserMsgId = entry.uuid;
+
+        if (!sessionToFirstUserMsgId.has(entry.sessionId)) {
+          sessionToFirstUserMsgId.set(entry.sessionId, firstUserMsgId);
+
+          const session = allSessions.get(entry.sessionId);
+          if (session) {
+            if (!sessionGroups.has(firstUserMsgId)) {
+              sessionGroups.set(firstUserMsgId, {
+                latestSession: session,
+                allSessions: [session]
+              });
+            } else {
+              const group = sessionGroups.get(firstUserMsgId);
+              group.allSessions.push(session);
+
+              // Update latest session if this one is more recent
+              if (new Date(session.lastActivity) > new Date(group.latestSession.lastActivity)) {
+                group.latestSession = session;
+              }
+            }
+          }
+        }
+      }
+    });
+
+    // Collect all sessions that don't belong to any group (standalone sessions)
+    const groupedSessionIds = new Set();
+    sessionGroups.forEach(group => {
+      group.allSessions.forEach(session => groupedSessionIds.add(session.id));
+    });
+
+    const standaloneSessionsArray = Array.from(allSessions.values())
+      .filter(session => !groupedSessionIds.has(session.id));
+
+    // Combine grouped sessions (only show latest from each group) + standalone sessions
+    const latestFromGroups = Array.from(sessionGroups.values()).map(group => {
+      const session = { ...group.latestSession };
+      // Add metadata about grouping
+      if (group.allSessions.length > 1) {
+        session.isGrouped = true;
+        session.groupSize = group.allSessions.length;
+        session.groupSessions = group.allSessions.map(s => s.id);
+      }
+      return session;
+    });
+    const visibleSessions = [...latestFromGroups, ...standaloneSessionsArray]
+      .filter(session => !session.summary.startsWith('{ "'))
+      .sort((a, b) => new Date(b.lastActivity) - new Date(a.lastActivity));
+
+    const total = visibleSessions.length;
+    const paginatedSessions = visibleSessions.slice(offset, offset + limit);
+    const hasMore = offset + limit < total;
+
+    return {
+      sessions: paginatedSessions,
+      hasMore,
+      total,
+      offset,
+      limit
+    };
+  } catch (error) {
+    console.error(`Error reading sessions for project ${projectName}:`, error);
+    return { sessions: [], hasMore: false, total: 0 };
+  }
+}
+
+async function parseJsonlSessions(filePath) {
+  const sessions = new Map();
+  const entries = [];
+  const pendingSummaries = new Map(); // leafUuid -> summary for entries without sessionId
+
+  try {
+    const fileStream = fsSync.createReadStream(filePath);
+    const rl = readline.createInterface({
+      input: fileStream,
+      crlfDelay: Infinity
+    });
+
+    for await (const line of rl) {
+      if (line.trim()) {
+        try {
+          const entry = JSON.parse(line);
+          entries.push(entry);
+
+          // Handle summary entries that don't have sessionId yet
+          if (entry.type === 'summary' && entry.summary && !entry.sessionId && entry.leafUuid) {
+            pendingSummaries.set(entry.leafUuid, entry.summary);
+          }
+
+          if (entry.sessionId) {
+            if (!sessions.has(entry.sessionId)) {
+              sessions.set(entry.sessionId, {
+                id: entry.sessionId,
+                summary: 'New Session',
+                messageCount: 0,
+                lastActivity: new Date(),
+                cwd: entry.cwd || '',
+                lastUserMessage: null,
+                lastAssistantMessage: null
+              });
+            }
+
+            const session = sessions.get(entry.sessionId);
+
+            // Apply pending summary if this entry has a parentUuid that matches a pending summary
+            if (session.summary === 'New Session' && entry.parentUuid && pendingSummaries.has(entry.parentUuid)) {
+              session.summary = pendingSummaries.get(entry.parentUuid);
+            }
+
+            // Update summary from summary entries with sessionId
+            if (entry.type === 'summary' && entry.summary) {
+              session.summary = entry.summary;
+            }
+
+            // Track last user and assistant messages (skip system messages)
+            if (entry.message?.role === 'user' && entry.message?.content) {
+              const content = entry.message.content;
+
+              // Extract text from array format if needed
+              let textContent = content;
+              if (Array.isArray(content) && content.length > 0 && content[0].type === 'text') {
+                textContent = content[0].text;
+              }
+
+              const isSystemMessage = typeof textContent === 'string' && (
+                textContent.startsWith('<command-name>') ||
+                textContent.startsWith('<command-message>') ||
+                textContent.startsWith('<command-args>') ||
+                textContent.startsWith('<local-command-stdout>') ||
+                textContent.startsWith('<system-reminder>') ||
+                textContent.startsWith('Caveat:') ||
+                textContent.startsWith('This session is being continued from a previous') ||
+                textContent.startsWith('Invalid API key') ||
+                textContent.includes('{"subtasks":') || // Filter Task Master prompts
+                textContent.includes('CRITICAL: You MUST respond with ONLY a JSON') || // Filter Task Master system prompts
+                textContent === 'Warmup' // Explicitly filter out "Warmup"
+              );
+
+              if (typeof textContent === 'string' && textContent.length > 0 && !isSystemMessage) {
+                session.lastUserMessage = textContent;
+              }
+            } else if (entry.message?.role === 'assistant' && entry.message?.content) {
+              // Skip API error messages using the isApiErrorMessage flag
+              if (entry.isApiErrorMessage === true) {
+                // Skip this message entirely
+              } else {
+                // Track last assistant text message
+                let assistantText = null;
+
+                if (Array.isArray(entry.message.content)) {
+                  for (const part of entry.message.content) {
+                    if (part.type === 'text' && part.text) {
+                      assistantText = part.text;
+                    }
+                  }
+                } else if (typeof entry.message.content === 'string') {
+                  assistantText = entry.message.content;
+                }
+
+                // Additional filter for assistant messages with system content
+                const isSystemAssistantMessage = typeof assistantText === 'string' && (
+                  assistantText.startsWith('Invalid API key') ||
+                  assistantText.includes('{"subtasks":') ||
+                  assistantText.includes('CRITICAL: You MUST respond with ONLY a JSON')
+                );
+
+                if (assistantText && !isSystemAssistantMessage) {
+                  session.lastAssistantMessage = assistantText;
+                }
+              }
+            }
+
+            session.messageCount++;
+
+            if (entry.timestamp) {
+              session.lastActivity = new Date(entry.timestamp);
+            }
+          }
+        } catch (parseError) {
+          // Skip malformed lines silently
+        }
+      }
+    }
+
+    // After processing all entries, set final summary based on last message if no summary exists
+    for (const session of sessions.values()) {
+      if (session.summary === 'New Session') {
+        // Prefer last user message, fall back to last assistant message
+        const lastMessage = session.lastUserMessage || session.lastAssistantMessage;
+        if (lastMessage) {
+          session.summary = lastMessage.length > 50 ? lastMessage.substring(0, 50) + '...' : lastMessage;
+        }
+      }
+    }
+
+    // Filter out sessions that contain JSON responses (Task Master errors)
+    const allSessions = Array.from(sessions.values());
+    const filteredSessions = allSessions.filter(session => {
+      const shouldFilter = session.summary.startsWith('{ "');
+      if (shouldFilter) {
+      }
+      // Log a sample of summaries to debug
+      if (Math.random() < 0.01) { // Log 1% of sessions
+      }
+      return !shouldFilter;
+    });
+
+
+    return {
+      sessions: filteredSessions,
+      entries: entries
+    };
+
+  } catch (error) {
+    console.error('Error reading JSONL file:', error);
+    return { sessions: [], entries: [] };
+  }
+}
+
+// Parse an agent JSONL file and extract tool uses
+async function parseAgentTools(filePath) {
+  const tools = [];
+
+  try {
+    const fileStream = fsSync.createReadStream(filePath);
+    const rl = readline.createInterface({
+      input: fileStream,
+      crlfDelay: Infinity
+    });
+
+    for await (const line of rl) {
+      if (line.trim()) {
+        try {
+          const entry = JSON.parse(line);
+          // Look for assistant messages with tool_use
+          if (entry.message?.role === 'assistant' && Array.isArray(entry.message?.content)) {
+            for (const part of entry.message.content) {
+              if (part.type === 'tool_use') {
+                tools.push({
+                  toolId: part.id,
+                  toolName: part.name,
+                  toolInput: part.input,
+                  timestamp: entry.timestamp
+                });
+              }
+            }
+          }
+          // Look for tool results
+          if (entry.message?.role === 'user' && Array.isArray(entry.message?.content)) {
+            for (const part of entry.message.content) {
+              if (part.type === 'tool_result') {
+                // Find the matching tool and add result
+                const tool = tools.find(t => t.toolId === part.tool_use_id);
+                if (tool) {
+                  tool.toolResult = {
+                    content: typeof part.content === 'string' ? part.content :
+                      Array.isArray(part.content) ? part.content.map(c => c.text || '').join('\n') :
+                        JSON.stringify(part.content),
+                    isError: Boolean(part.is_error)
+                  };
+                }
+              }
+            }
+          }
+        } catch (parseError) {
+          // Skip malformed lines
+        }
+      }
+    }
+  } catch (error) {
+    console.warn(`Error parsing agent file ${filePath}:`, error.message);
+  }
+
+  return tools;
+}
+
+// Get messages for a specific session with pagination support
+async function getSessionMessages(projectName, sessionId, limit = null, offset = 0) {
+  const projectDir = path.join(os.homedir(), '.claude', 'projects', projectName);
+
+  try {
+    const files = await fs.readdir(projectDir);
+    // agent-*.jsonl files contain subagent tool history - we'll process them separately
+    const jsonlFiles = files.filter(file => file.endsWith('.jsonl') && !file.startsWith('agent-'));
+    const agentFiles = files.filter(file => file.endsWith('.jsonl') && file.startsWith('agent-'));
+
+    if (jsonlFiles.length === 0) {
+      return { messages: [], total: 0, hasMore: false };
+    }
+
+    const messages = [];
+    // Map of agentId -> tools for subagent tool grouping
+    const agentToolsCache = new Map();
+
+    // Process all JSONL files to find messages for this session
+    for (const file of jsonlFiles) {
+      const jsonlFile = path.join(projectDir, file);
+      const fileStream = fsSync.createReadStream(jsonlFile);
+      const rl = readline.createInterface({
+        input: fileStream,
+        crlfDelay: Infinity
+      });
+
+      for await (const line of rl) {
+        if (line.trim()) {
+          try {
+            const entry = JSON.parse(line);
+            if (entry.sessionId === sessionId) {
+              messages.push(entry);
+            }
+          } catch (parseError) {
+            // Silently skip malformed JSONL lines (common with concurrent writes)
+          }
+        }
+      }
+    }
+
+    // Collect agentIds from Task tool results
+    const agentIds = new Set();
+    for (const message of messages) {
+      if (message.toolUseResult?.agentId) {
+        agentIds.add(message.toolUseResult.agentId);
+      }
+    }
+
+    // Load agent tools for each agentId found
+    for (const agentId of agentIds) {
+      const agentFileName = `agent-${agentId}.jsonl`;
+      if (agentFiles.includes(agentFileName)) {
+        const agentFilePath = path.join(projectDir, agentFileName);
+        const tools = await parseAgentTools(agentFilePath);
+        agentToolsCache.set(agentId, tools);
+      }
+    }
+
+    // Attach agent tools to their parent Task messages
+    for (const message of messages) {
+      if (message.toolUseResult?.agentId) {
+        const agentId = message.toolUseResult.agentId;
+        const agentTools = agentToolsCache.get(agentId);
+        if (agentTools && agentTools.length > 0) {
+          message.subagentTools = agentTools;
+        }
+      }
+    }
+    // Sort messages by timestamp
+    const sortedMessages = messages.sort((a, b) =>
+      new Date(a.timestamp || 0) - new Date(b.timestamp || 0)
+    );
+
+    const total = sortedMessages.length;
+
+    // If no limit is specified, return all messages (backward compatibility)
+    if (limit === null) {
+      return sortedMessages;
+    }
+
+    // Apply pagination - for recent messages, we need to slice from the end
+    // offset 0 should give us the most recent messages
+    const startIndex = Math.max(0, total - offset - limit);
+    const endIndex = total - offset;
+    const paginatedMessages = sortedMessages.slice(startIndex, endIndex);
+    const hasMore = startIndex > 0;
+
+    return {
+      messages: paginatedMessages,
+      total,
+      hasMore,
+      offset,
+      limit
+    };
+  } catch (error) {
+    console.error(`Error reading messages for session ${sessionId}:`, error);
+    return limit === null ? [] : { messages: [], total: 0, hasMore: false };
+  }
+}
+
+// Rename a project's display name
+async function renameProject(projectName, newDisplayName) {
+  const config = await loadProjectConfig();
+
+  if (!newDisplayName || newDisplayName.trim() === '') {
+    // Remove custom name if empty, will fall back to auto-generated
+    if (config[projectName]) {
+      delete config[projectName].displayName;
+    }
+  } else {
+    // Set custom display name, preserving other properties (manuallyAdded, originalPath)
+    config[projectName] = {
+      ...config[projectName],
+      displayName: newDisplayName.trim()
+    };
+  }
+
+  await saveProjectConfig(config);
+  return true;
+}
+
+// Delete a session from a project
+async function deleteSession(projectName, sessionId) {
+  const projectDir = path.join(os.homedir(), '.claude', 'projects', projectName);
+
+  try {
+    const files = await fs.readdir(projectDir);
+    const jsonlFiles = files.filter(file => file.endsWith('.jsonl'));
+
+    if (jsonlFiles.length === 0) {
+      throw new Error('No session files found for this project');
+    }
+
+    // Check all JSONL files to find which one contains the session
+    for (const file of jsonlFiles) {
+      const jsonlFile = path.join(projectDir, file);
+      const content = await fs.readFile(jsonlFile, 'utf8');
+      const lines = content.split('\n').filter(line => line.trim());
+
+      // Check if this file contains the session
+      const hasSession = lines.some(line => {
+        try {
+          const data = JSON.parse(line);
+          return data.sessionId === sessionId;
+        } catch {
+          return false;
+        }
+      });
+
+      if (hasSession) {
+        // Filter out all entries for this session
+        const filteredLines = lines.filter(line => {
+          try {
+            const data = JSON.parse(line);
+            return data.sessionId !== sessionId;
+          } catch {
+            return true; // Keep malformed lines
+          }
+        });
+
+        // Write back the filtered content
+        await fs.writeFile(jsonlFile, filteredLines.join('\n') + (filteredLines.length > 0 ? '\n' : ''));
+        return true;
+      }
+    }
+
+    throw new Error(`Session ${sessionId} not found in any files`);
+  } catch (error) {
+    console.error(`Error deleting session ${sessionId} from project ${projectName}:`, error);
+    throw error;
+  }
+}
+
+// Check if a project is empty (has no sessions)
+async function isProjectEmpty(projectName) {
+  try {
+    const sessionsResult = await getSessions(projectName, 1, 0);
+    return sessionsResult.total === 0;
+  } catch (error) {
+    console.error(`Error checking if project ${projectName} is empty:`, error);
+    return false;
+  }
+}
+
+// Delete a project (force=true to delete even with sessions)
+async function deleteProject(projectName, force = false) {
+  const projectDir = path.join(os.homedir(), '.claude', 'projects', projectName);
+
+  try {
+    const isEmpty = await isProjectEmpty(projectName);
+    if (!isEmpty && !force) {
+      throw new Error('Cannot delete project with existing sessions');
+    }
+
+    const config = await loadProjectConfig();
+    let projectPath = config[projectName]?.path || config[projectName]?.originalPath;
+
+    // Fallback to extractProjectDirectory if projectPath is not in config
+    if (!projectPath) {
+      projectPath = await extractProjectDirectory(projectName);
+    }
+
+    // Remove the project directory (includes all Claude sessions)
+    await fs.rm(projectDir, { recursive: true, force: true });
+
+    // Delete all Codex sessions associated with this project
+    if (projectPath) {
+      try {
+        const codexSessions = await getCodexSessions(projectPath, { limit: 0 });
+        for (const session of codexSessions) {
+          try {
+            await deleteCodexSession(session.id);
+          } catch (err) {
+            console.warn(`Failed to delete Codex session ${session.id}:`, err.message);
+          }
+        }
+      } catch (err) {
+        console.warn('Failed to delete Codex sessions:', err.message);
+      }
+
+      // Delete Cursor sessions directory if it exists
+      try {
+        const hash = crypto.createHash('md5').update(projectPath).digest('hex');
+        const cursorProjectDir = path.join(os.homedir(), '.cursor', 'chats', hash);
+        await fs.rm(cursorProjectDir, { recursive: true, force: true });
+      } catch (err) {
+        // Cursor dir may not exist, ignore
+      }
+    }
+
+    // Remove from project config
+    delete config[projectName];
+    await saveProjectConfig(config);
+
+    return true;
+  } catch (error) {
+    console.error(`Error deleting project ${projectName}:`, error);
+    throw error;
+  }
+}
+
+// Add a project manually to the config (without creating folders)
+async function addProjectManually(projectPath, displayName = null) {
+  const absolutePath = path.resolve(projectPath);
+
+  try {
+    // Check if the path exists
+    await fs.access(absolutePath);
+  } catch (error) {
+    throw new Error(`Path does not exist: ${absolutePath}`);
+  }
+
+  // Generate project name (encode path for use as directory name)
+  const projectName = absolutePath.replace(/[\\/:\s~_]/g, '-');
+
+  // Check if project already exists in config
+  const config = await loadProjectConfig();
+  const projectDir = path.join(os.homedir(), '.claude', 'projects', projectName);
+
+  if (config[projectName]) {
+    throw new Error(`Project already configured for path: ${absolutePath}`);
+  }
+
+  // Allow adding projects even if the directory exists - this enables tracking
+  // existing Claude Code or Cursor projects in the UI
+
+  // Add to config as manually added project
+  config[projectName] = {
+    manuallyAdded: true,
+    originalPath: absolutePath
+  };
+
+  if (displayName) {
+    config[projectName].displayName = displayName;
+  }
+
+  await saveProjectConfig(config);
+
+
+  return {
+    name: projectName,
+    path: absolutePath,
+    fullPath: absolutePath,
+    displayName: displayName || await generateDisplayName(projectName, absolutePath),
+    isManuallyAdded: true,
+    sessions: [],
+    cursorSessions: []
+  };
+}
+
+// Fetch Cursor sessions for a given project path
+async function getCursorSessions(projectPath) {
+  try {
+    // Calculate cwdID hash for the project path (Cursor uses MD5 hash)
+    const cwdId = crypto.createHash('md5').update(projectPath).digest('hex');
+    const cursorChatsPath = path.join(os.homedir(), '.cursor', 'chats', cwdId);
+
+    // Check if the directory exists
+    try {
+      await fs.access(cursorChatsPath);
+    } catch (error) {
+      // No sessions for this project
+      return [];
+    }
+
+    // List all session directories
+    const sessionDirs = await fs.readdir(cursorChatsPath);
+    const sessions = [];
+
+    for (const sessionId of sessionDirs) {
+      const sessionPath = path.join(cursorChatsPath, sessionId);
+      const storeDbPath = path.join(sessionPath, 'store.db');
+
+      try {
+        // Check if store.db exists
+        await fs.access(storeDbPath);
+
+        // Capture store.db mtime as a reliable fallback timestamp
+        let dbStatMtimeMs = null;
+        try {
+          const stat = await fs.stat(storeDbPath);
+          dbStatMtimeMs = stat.mtimeMs;
+        } catch (_) { }
+
+        // Open SQLite database
+        const db = await open({
+          filename: storeDbPath,
+          driver: sqlite3.Database,
+          mode: sqlite3.OPEN_READONLY
+        });
+
+        // Get metadata from meta table
+        const metaRows = await db.all(`
+          SELECT key, value FROM meta
+        `);
+
+        // Parse metadata
+        let metadata = {};
+        for (const row of metaRows) {
+          if (row.value) {
+            try {
+              // Try to decode as hex-encoded JSON
+              const hexMatch = row.value.toString().match(/^[0-9a-fA-F]+$/);
+              if (hexMatch) {
+                const jsonStr = Buffer.from(row.value, 'hex').toString('utf8');
+                metadata[row.key] = JSON.parse(jsonStr);
+              } else {
+                metadata[row.key] = row.value.toString();
+              }
+            } catch (e) {
+              metadata[row.key] = row.value.toString();
+            }
+          }
+        }
+
+        // Get message count
+        const messageCountResult = await db.get(`
+          SELECT COUNT(*) as count FROM blobs
+        `);
+
+        await db.close();
+
+        // Extract session info
+        const sessionName = metadata.title || metadata.sessionTitle || 'Untitled Session';
+
+        // Determine timestamp - prefer createdAt from metadata, fall back to db file mtime
+        let createdAt = null;
+        if (metadata.createdAt) {
+          createdAt = new Date(metadata.createdAt).toISOString();
+        } else if (dbStatMtimeMs) {
+          createdAt = new Date(dbStatMtimeMs).toISOString();
+        } else {
+          createdAt = new Date().toISOString();
+        }
+
+        sessions.push({
+          id: sessionId,
+          name: sessionName,
+          createdAt: createdAt,
+          lastActivity: createdAt, // For compatibility with Claude sessions
+          messageCount: messageCountResult.count || 0,
+          projectPath: projectPath
+        });
+
+      } catch (error) {
+        console.warn(`Could not read Cursor session ${sessionId}:`, error.message);
+      }
+    }
+
+    // Sort sessions by creation time (newest first)
+    sessions.sort((a, b) => new Date(b.createdAt) - new Date(a.createdAt));
+
+    // Return only the first 5 sessions for performance
+    return sessions.slice(0, 5);
+
+  } catch (error) {
+    console.error('Error fetching Cursor sessions:', error);
+    return [];
+  }
+}
+
+
+function normalizeComparablePath(inputPath) {
+  if (!inputPath || typeof inputPath !== 'string') {
+    return '';
+  }
+
+  const withoutLongPathPrefix = inputPath.startsWith('\\\\?\\')
+    ? inputPath.slice(4)
+    : inputPath;
+  const normalized = path.normalize(withoutLongPathPrefix.trim());
+
+  if (!normalized) {
+    return '';
+  }
+
+  const resolved = path.resolve(normalized);
+  return process.platform === 'win32' ? resolved.toLowerCase() : resolved;
+}
+
+async function findCodexJsonlFiles(dir) {
+  const files = [];
+
+  try {
+    const entries = await fs.readdir(dir, { withFileTypes: true });
+    for (const entry of entries) {
+      const fullPath = path.join(dir, entry.name);
+      if (entry.isDirectory()) {
+        files.push(...await findCodexJsonlFiles(fullPath));
+      } else if (entry.name.endsWith('.jsonl')) {
+        files.push(fullPath);
+      }
+    }
+  } catch (error) {
+    // Skip directories we can't read
+  }
+
+  return files;
+}
+
+async function buildCodexSessionsIndex() {
+  const codexSessionsDir = path.join(os.homedir(), '.codex', 'sessions');
+  const sessionsByProject = new Map();
+
+  try {
+    await fs.access(codexSessionsDir);
+  } catch (error) {
+    return sessionsByProject;
+  }
+
+  const jsonlFiles = await findCodexJsonlFiles(codexSessionsDir);
+
+  for (const filePath of jsonlFiles) {
+    try {
+      const sessionData = await parseCodexSessionFile(filePath);
+      if (!sessionData || !sessionData.id) {
+        continue;
+      }
+
+      const normalizedProjectPath = normalizeComparablePath(sessionData.cwd);
+      if (!normalizedProjectPath) {
+        continue;
+      }
+
+      const session = {
+        id: sessionData.id,
+        summary: sessionData.summary || 'Codex Session',
+        messageCount: sessionData.messageCount || 0,
+        lastActivity: sessionData.timestamp ? new Date(sessionData.timestamp) : new Date(),
+        cwd: sessionData.cwd,
+        model: sessionData.model,
+        filePath,
+        provider: 'codex',
+      };
+
+      if (!sessionsByProject.has(normalizedProjectPath)) {
+        sessionsByProject.set(normalizedProjectPath, []);
+      }
+
+      sessionsByProject.get(normalizedProjectPath).push(session);
+    } catch (error) {
+      console.warn(`Could not parse Codex session file ${filePath}:`, error.message);
+    }
+  }
+
+  for (const sessions of sessionsByProject.values()) {
+    sessions.sort((a, b) => new Date(b.lastActivity) - new Date(a.lastActivity));
+  }
+
+  return sessionsByProject;
+}
+
+// Fetch Codex sessions for a given project path
+async function getCodexSessions(projectPath, options = {}) {
+  const { limit = 5, indexRef = null } = options;
+  try {
+    const normalizedProjectPath = normalizeComparablePath(projectPath);
+    if (!normalizedProjectPath) {
+      return [];
+    }
+
+    if (indexRef && !indexRef.sessionsByProject) {
+      indexRef.sessionsByProject = await buildCodexSessionsIndex();
+    }
+
+    const sessionsByProject = indexRef?.sessionsByProject || await buildCodexSessionsIndex();
+    const sessions = sessionsByProject.get(normalizedProjectPath) || [];
+
+    // Return limited sessions for performance (0 = unlimited for deletion)
+    return limit > 0 ? sessions.slice(0, limit) : [...sessions];
+
+  } catch (error) {
+    console.error('Error fetching Codex sessions:', error);
+    return [];
+  }
+}
+
+function isVisibleCodexUserMessage(payload) {
+  if (!payload || payload.type !== 'user_message') {
+    return false;
+  }
+
+  // Codex logs internal context (environment, instructions) as non-plain user_message kinds.
+  if (payload.kind && payload.kind !== 'plain') {
+    return false;
+  }
+
+  if (typeof payload.message !== 'string' || payload.message.trim().length === 0) {
+    return false;
+  }
+  
+  return true;
+}
+
+// Parse a Codex session JSONL file to extract metadata
+async function parseCodexSessionFile(filePath) {
+  try {
+    const fileStream = fsSync.createReadStream(filePath);
+    const rl = readline.createInterface({
+      input: fileStream,
+      crlfDelay: Infinity
+    });
+
+    let sessionMeta = null;
+    let lastTimestamp = null;
+    let lastUserMessage = null;
+    let messageCount = 0;
+
+    for await (const line of rl) {
+      if (line.trim()) {
+        try {
+          const entry = JSON.parse(line);
+
+          // Track timestamp
+          if (entry.timestamp) {
+            lastTimestamp = entry.timestamp;
+          }
+
+          // Extract session metadata
+          if (entry.type === 'session_meta' && entry.payload) {
+            sessionMeta = {
+              id: entry.payload.id,
+              cwd: entry.payload.cwd,
+              model: entry.payload.model || entry.payload.model_provider,
+              timestamp: entry.timestamp,
+              git: entry.payload.git
+            };
+          }
+
+          // Count visible user messages and extract summary from the latest plain user input.
+          if (entry.type === 'event_msg' && isVisibleCodexUserMessage(entry.payload)) {
+            messageCount++;
+            if (entry.payload.message) {
+              lastUserMessage = entry.payload.message;
+            }
+          }
+
+          if (entry.type === 'response_item' && entry.payload?.type === 'message' && entry.payload.role === 'assistant') {
+            messageCount++;
+          }
+
+        } catch (parseError) {
+          // Skip malformed lines
+        }
+      }
+    }
+
+    if (sessionMeta) {
+      return {
+        ...sessionMeta,
+        timestamp: lastTimestamp || sessionMeta.timestamp,
+        summary: lastUserMessage ?
+          (lastUserMessage.length > 50 ? lastUserMessage.substring(0, 50) + '...' : lastUserMessage) :
+          'Codex Session',
+        messageCount
+      };
+    }
+
+    return null;
+
+  } catch (error) {
+    console.error('Error parsing Codex session file:', error);
+    return null;
+  }
+}
+
+// Get messages for a specific Codex session
+async function getCodexSessionMessages(sessionId, limit = null, offset = 0) {
+  try {
+    const codexSessionsDir = path.join(os.homedir(), '.codex', 'sessions');
+
+    // Find the session file by searching for the session ID
+    const findSessionFile = async (dir) => {
+      try {
+        const entries = await fs.readdir(dir, { withFileTypes: true });
+        for (const entry of entries) {
+          const fullPath = path.join(dir, entry.name);
+          if (entry.isDirectory()) {
+            const found = await findSessionFile(fullPath);
+            if (found) return found;
+          } else if (entry.name.includes(sessionId) && entry.name.endsWith('.jsonl')) {
+            return fullPath;
+          }
+        }
+      } catch (error) {
+        // Skip directories we can't read
+      }
+      return null;
+    };
+
+    const sessionFilePath = await findSessionFile(codexSessionsDir);
+
+    if (!sessionFilePath) {
+      console.warn(`Codex session file not found for session ${sessionId}`);
+      return { messages: [], total: 0, hasMore: false };
+    }
+
+    const messages = [];
+    let tokenUsage = null;
+    const fileStream = fsSync.createReadStream(sessionFilePath);
+    const rl = readline.createInterface({
+      input: fileStream,
+      crlfDelay: Infinity
+    });
+
+    // Helper to extract text from Codex content array
+    const extractText = (content) => {
+      if (!Array.isArray(content)) return content;
+      return content
+        .map(item => {
+          if (item.type === 'input_text' || item.type === 'output_text') {
+            return item.text;
+          }
+          if (item.type === 'text') {
+            return item.text;
+          }
+          return '';
+        })
+        .filter(Boolean)
+        .join('\n');
+    };
+
+    for await (const line of rl) {
+      if (line.trim()) {
+        try {
+          const entry = JSON.parse(line);
+
+          // Extract token usage from token_count events (keep latest)
+          if (entry.type === 'event_msg' && entry.payload?.type === 'token_count' && entry.payload?.info) {
+            const info = entry.payload.info;
+            if (info.total_token_usage) {
+              tokenUsage = {
+                used: info.total_token_usage.total_tokens || 0,
+                total: info.model_context_window || 200000
+              };
+            }
+          }
+          
+          // Use event_msg.user_message for user-visible inputs.
+          if (entry.type === 'event_msg' && isVisibleCodexUserMessage(entry.payload)) {
+            messages.push({
+              type: 'user',
+              timestamp: entry.timestamp,
+              message: {
+                role: 'user',
+                content: entry.payload.message
+              }
+            });
+          }
+
+          // response_item.message may include internal prompts for non-assistant roles.
+          // Keep only assistant output from response_item.
+          if (
+            entry.type === 'response_item' &&
+            entry.payload?.type === 'message' &&
+            entry.payload.role === 'assistant'
+          ) {
+            const content = entry.payload.content;
+            const textContent = extractText(content);
+
+            // Only add if there's actual content
+            if (textContent?.trim()) {
+              messages.push({
+                type: 'assistant',
+                timestamp: entry.timestamp,
+                message: {
+                  role: 'assistant',
+                  content: textContent
+                }
+              });
+            }
+          }
+
+          if (entry.type === 'response_item' && entry.payload?.type === 'reasoning') {
+            const summaryText = entry.payload.summary
+              ?.map(s => s.text)
+              .filter(Boolean)
+              .join('\n');
+            if (summaryText?.trim()) {
+              messages.push({
+                type: 'thinking',
+                timestamp: entry.timestamp,
+                message: {
+                  role: 'assistant',
+                  content: summaryText
+                }
+              });
+            }
+          }
+
+          if (entry.type === 'response_item' && entry.payload?.type === 'function_call') {
+            let toolName = entry.payload.name;
+            let toolInput = entry.payload.arguments;
+
+            // Map Codex tool names to Claude equivalents
+            if (toolName === 'shell_command') {
+              toolName = 'Bash';
+              try {
+                const args = JSON.parse(entry.payload.arguments);
+                toolInput = JSON.stringify({ command: args.command });
+              } catch (e) {
+                // Keep original if parsing fails
+              }
+            }
+
+            messages.push({
+              type: 'tool_use',
+              timestamp: entry.timestamp,
+              toolName: toolName,
+              toolInput: toolInput,
+              toolCallId: entry.payload.call_id
+            });
+          }
+
+          if (entry.type === 'response_item' && entry.payload?.type === 'function_call_output') {
+            messages.push({
+              type: 'tool_result',
+              timestamp: entry.timestamp,
+              toolCallId: entry.payload.call_id,
+              output: entry.payload.output
+            });
+          }
+
+          if (entry.type === 'response_item' && entry.payload?.type === 'custom_tool_call') {
+            const toolName = entry.payload.name || 'custom_tool';
+            const input = entry.payload.input || '';
+
+            if (toolName === 'apply_patch') {
+              // Parse Codex patch format and convert to Claude Edit format
+              const fileMatch = input.match(/\*\*\* Update File: (.+)/);
+              const filePath = fileMatch ? fileMatch[1].trim() : 'unknown';
+
+              // Extract old and new content from patch
+              const lines = input.split('\n');
+              const oldLines = [];
+              const newLines = [];
+
+              for (const line of lines) {
+                if (line.startsWith('-') && !line.startsWith('---')) {
+                  oldLines.push(line.substring(1));
+                } else if (line.startsWith('+') && !line.startsWith('+++')) {
+                  newLines.push(line.substring(1));
+                }
+              }
+
+              messages.push({
+                type: 'tool_use',
+                timestamp: entry.timestamp,
+                toolName: 'Edit',
+                toolInput: JSON.stringify({
+                  file_path: filePath,
+                  old_string: oldLines.join('\n'),
+                  new_string: newLines.join('\n')
+                }),
+                toolCallId: entry.payload.call_id
+              });
+            } else {
+              messages.push({
+                type: 'tool_use',
+                timestamp: entry.timestamp,
+                toolName: toolName,
+                toolInput: input,
+                toolCallId: entry.payload.call_id
+              });
+            }
+          }
+
+          if (entry.type === 'response_item' && entry.payload?.type === 'custom_tool_call_output') {
+            messages.push({
+              type: 'tool_result',
+              timestamp: entry.timestamp,
+              toolCallId: entry.payload.call_id,
+              output: entry.payload.output || ''
+            });
+          }
+
+        } catch (parseError) {
+          // Skip malformed lines
+        }
+      }
+    }
+
+    // Sort by timestamp
+    messages.sort((a, b) => new Date(a.timestamp || 0) - new Date(b.timestamp || 0));
+
+    const total = messages.length;
+
+    // Apply pagination if limit is specified
+    if (limit !== null) {
+      const startIndex = Math.max(0, total - offset - limit);
+      const endIndex = total - offset;
+      const paginatedMessages = messages.slice(startIndex, endIndex);
+      const hasMore = startIndex > 0;
+
+      return {
+        messages: paginatedMessages,
+        total,
+        hasMore,
+        offset,
+        limit,
+        tokenUsage
+      };
+    }
+
+    return { messages, tokenUsage };
+
+  } catch (error) {
+    console.error(`Error reading Codex session messages for ${sessionId}:`, error);
+    return { messages: [], total: 0, hasMore: false };
+  }
+}
+
+async function deleteCodexSession(sessionId) {
+  try {
+    const codexSessionsDir = path.join(os.homedir(), '.codex', 'sessions');
+
+    const findJsonlFiles = async (dir) => {
+      const files = [];
+      try {
+        const entries = await fs.readdir(dir, { withFileTypes: true });
+        for (const entry of entries) {
+          const fullPath = path.join(dir, entry.name);
+          if (entry.isDirectory()) {
+            files.push(...await findJsonlFiles(fullPath));
+          } else if (entry.name.endsWith('.jsonl')) {
+            files.push(fullPath);
+          }
+        }
+      } catch (error) { }
+      return files;
+    };
+
+    const jsonlFiles = await findJsonlFiles(codexSessionsDir);
+
+    for (const filePath of jsonlFiles) {
+      const sessionData = await parseCodexSessionFile(filePath);
+      if (sessionData && sessionData.id === sessionId) {
+        await fs.unlink(filePath);
+        return true;
+      }
+    }
+
+    throw new Error(`Codex session file not found for session ${sessionId}`);
+  } catch (error) {
+    console.error(`Error deleting Codex session ${sessionId}:`, error);
+    throw error;
+  }
+}
+
+async function searchConversations(query, limit = 50, onProjectResult = null, signal = null) {
+  const safeQuery = typeof query === 'string' ? query.trim() : '';
+  const safeLimit = Math.max(1, Math.min(Number.isFinite(limit) ? limit : 50, 200));
+  const claudeDir = path.join(os.homedir(), '.claude', 'projects');
+  const config = await loadProjectConfig();
+  const results = [];
+  let totalMatches = 0;
+  const words = safeQuery.toLowerCase().split(/\s+/).filter(w => w.length > 0);
+  if (words.length === 0) return { results: [], totalMatches: 0, query: safeQuery };
+
+  const isAborted = () => signal?.aborted === true;
+
+  const isSystemMessage = (textContent) => {
+    return typeof textContent === 'string' && (
+      textContent.startsWith('<command-name>') ||
+      textContent.startsWith('<command-message>') ||
+      textContent.startsWith('<command-args>') ||
+      textContent.startsWith('<local-command-stdout>') ||
+      textContent.startsWith('<system-reminder>') ||
+      textContent.startsWith('Caveat:') ||
+      textContent.startsWith('This session is being continued from a previous') ||
+      textContent.startsWith('Invalid API key') ||
+      textContent.includes('{"subtasks":') ||
+      textContent.includes('CRITICAL: You MUST respond with ONLY a JSON') ||
+      textContent === 'Warmup'
+    );
+  };
+
+  const extractText = (content) => {
+    if (typeof content === 'string') return content;
+    if (Array.isArray(content)) {
+      return content
+        .filter(part => part.type === 'text' && part.text)
+        .map(part => part.text)
+        .join(' ');
+    }
+    return '';
+  };
+
+  const escapeRegex = (s) => s.replace(/[.*+?^${}()|[\]\\]/g, '\\$&');
+  const wordPatterns = words.map(w => new RegExp(`(?<!\\p{L})${escapeRegex(w)}(?!\\p{L})`, 'u'));
+  const allWordsMatch = (textLower) => {
+    return wordPatterns.every(p => p.test(textLower));
+  };
+
+  const buildSnippet = (text, textLower, snippetLen = 150) => {
+    let firstIndex = -1;
+    let firstWordLen = 0;
+    for (const w of words) {
+      const re = new RegExp(`(?<!\\p{L})${escapeRegex(w)}(?!\\p{L})`, 'u');
+      const m = re.exec(textLower);
+      if (m && (firstIndex === -1 || m.index < firstIndex)) {
+        firstIndex = m.index;
+        firstWordLen = w.length;
+      }
+    }
+    if (firstIndex === -1) firstIndex = 0;
+    const halfLen = Math.floor(snippetLen / 2);
+    let start = Math.max(0, firstIndex - halfLen);
+    let end = Math.min(text.length, firstIndex + halfLen + firstWordLen);
+    let snippet = text.slice(start, end).replace(/\n/g, ' ');
+    const prefix = start > 0 ? '...' : '';
+    const suffix = end < text.length ? '...' : '';
+    snippet = prefix + snippet + suffix;
+    const snippetLower = snippet.toLowerCase();
+    const highlights = [];
+    for (const word of words) {
+      const re = new RegExp(`(?<!\\p{L})${escapeRegex(word)}(?!\\p{L})`, 'gu');
+      let match;
+      while ((match = re.exec(snippetLower)) !== null) {
+        highlights.push({ start: match.index, end: match.index + word.length });
+      }
+    }
+    highlights.sort((a, b) => a.start - b.start);
+    const merged = [];
+    for (const h of highlights) {
+      const last = merged[merged.length - 1];
+      if (last && h.start <= last.end) {
+        last.end = Math.max(last.end, h.end);
+      } else {
+        merged.push({ ...h });
+      }
+    }
+    return { snippet, highlights: merged };
+  };
+
+  try {
+    await fs.access(claudeDir);
+    const entries = await fs.readdir(claudeDir, { withFileTypes: true });
+    const projectDirs = entries.filter(e => e.isDirectory());
+    let scannedProjects = 0;
+    const totalProjects = projectDirs.length;
+
+    for (const projectEntry of projectDirs) {
+      if (totalMatches >= safeLimit || isAborted()) break;
+
+      const projectName = projectEntry.name;
+      const projectDir = path.join(claudeDir, projectName);
+      const displayName = config[projectName]?.displayName
+        || await generateDisplayName(projectName);
+
+      let files;
+      try {
+        files = await fs.readdir(projectDir);
+      } catch {
+        continue;
+      }
+
+      const jsonlFiles = files.filter(
+        file => file.endsWith('.jsonl') && !file.startsWith('agent-')
+      );
+
+      const projectResult = {
+        projectName,
+        projectDisplayName: displayName,
+        sessions: []
+      };
+
+      for (const file of jsonlFiles) {
+        if (totalMatches >= safeLimit || isAborted()) break;
+
+        const filePath = path.join(projectDir, file);
+        const sessionMatches = new Map();
+        const sessionSummaries = new Map();
+        const pendingSummaries = new Map();
+        const sessionLastMessages = new Map();
+        let currentSessionId = null;
+
+        try {
+          const fileStream = fsSync.createReadStream(filePath);
+          const rl = readline.createInterface({
+            input: fileStream,
+            crlfDelay: Infinity
+          });
+
+          for await (const line of rl) {
+            if (totalMatches >= safeLimit || isAborted()) break;
+            if (!line.trim()) continue;
+
+            let entry;
+            try {
+              entry = JSON.parse(line);
+            } catch {
+              continue;
+            }
+
+            if (entry.sessionId) {
+              currentSessionId = entry.sessionId;
+            }
+            if (entry.type === 'summary' && entry.summary) {
+              const sid = entry.sessionId || currentSessionId;
+              if (sid) {
+                sessionSummaries.set(sid, entry.summary);
+              } else if (entry.leafUuid) {
+                pendingSummaries.set(entry.leafUuid, entry.summary);
+              }
+            }
+
+            // Apply pending summary via parentUuid
+            if (entry.parentUuid && currentSessionId && !sessionSummaries.has(currentSessionId)) {
+              const pending = pendingSummaries.get(entry.parentUuid);
+              if (pending) sessionSummaries.set(currentSessionId, pending);
+            }
+
+            // Track last user/assistant message for fallback title
+            if (entry.message?.content && currentSessionId && !entry.isApiErrorMessage) {
+              const role = entry.message.role;
+              if (role === 'user' || role === 'assistant') {
+                const text = extractText(entry.message.content);
+                if (text && !isSystemMessage(text)) {
+                  if (!sessionLastMessages.has(currentSessionId)) {
+                    sessionLastMessages.set(currentSessionId, {});
+                  }
+                  const msgs = sessionLastMessages.get(currentSessionId);
+                  if (role === 'user') msgs.user = text;
+                  else msgs.assistant = text;
+                }
+              }
+            }
+
+            if (!entry.message?.content) continue;
+            if (entry.message.role !== 'user' && entry.message.role !== 'assistant') continue;
+            if (entry.isApiErrorMessage) continue;
+
+            const text = extractText(entry.message.content);
+            if (!text || isSystemMessage(text)) continue;
+
+            const textLower = text.toLowerCase();
+            if (!allWordsMatch(textLower)) continue;
+
+            const sessionId = entry.sessionId || currentSessionId || file.replace('.jsonl', '');
+            if (!sessionMatches.has(sessionId)) {
+              sessionMatches.set(sessionId, []);
+            }
+
+            const matches = sessionMatches.get(sessionId);
+            if (matches.length < 2) {
+              const { snippet, highlights } = buildSnippet(text, textLower);
+              matches.push({
+                role: entry.message.role,
+                snippet,
+                highlights,
+                timestamp: entry.timestamp || null,
+                provider: 'claude',
+                messageUuid: entry.uuid || null
+              });
+              totalMatches++;
+            }
+          }
+        } catch {
+          continue;
+        }
+
+        for (const [sessionId, matches] of sessionMatches) {
+          projectResult.sessions.push({
+            sessionId,
+            provider: 'claude',
+            sessionSummary: sessionSummaries.get(sessionId) || (() => {
+              const msgs = sessionLastMessages.get(sessionId);
+              const lastMsg = msgs?.user || msgs?.assistant;
+              return lastMsg ? (lastMsg.length > 50 ? lastMsg.substring(0, 50) + '...' : lastMsg) : 'New Session';
+            })(),
+            matches
+          });
+        }
+      }
+
+      // Search Codex sessions for this project
+      try {
+        const actualProjectDir = await extractProjectDirectory(projectName);
+        if (actualProjectDir && !isAborted() && totalMatches < safeLimit) {
+          await searchCodexSessionsForProject(
+            actualProjectDir, projectResult, words, allWordsMatch, extractText, isSystemMessage,
+            buildSnippet, safeLimit, () => totalMatches, (n) => { totalMatches += n; }, isAborted
+          );
+        }
+      } catch {
+        // Skip codex search errors
+      }
+
+      // Search Gemini sessions for this project
+      try {
+        const actualProjectDir = await extractProjectDirectory(projectName);
+        if (actualProjectDir && !isAborted() && totalMatches < safeLimit) {
+          await searchGeminiSessionsForProject(
+            actualProjectDir, projectResult, words, allWordsMatch,
+            buildSnippet, safeLimit, () => totalMatches, (n) => { totalMatches += n; }
+          );
+        }
+      } catch {
+        // Skip gemini search errors
+      }
+
+      scannedProjects++;
+      if (projectResult.sessions.length > 0) {
+        results.push(projectResult);
+        if (onProjectResult) {
+          onProjectResult({ projectResult, totalMatches, scannedProjects, totalProjects });
+        }
+      } else if (onProjectResult && scannedProjects % 10 === 0) {
+        onProjectResult({ projectResult: null, totalMatches, scannedProjects, totalProjects });
+      }
+    }
+  } catch {
+    // claudeDir doesn't exist
+  }
+
+  return { results, totalMatches, query: safeQuery };
+}
+
+async function searchCodexSessionsForProject(
+  projectPath, projectResult, words, allWordsMatch, extractText, isSystemMessage,
+  buildSnippet, limit, getTotalMatches, addMatches, isAborted
+) {
+  const normalizedProjectPath = normalizeComparablePath(projectPath);
+  if (!normalizedProjectPath) return;
+  const codexSessionsDir = path.join(os.homedir(), '.codex', 'sessions');
+  try {
+    await fs.access(codexSessionsDir);
+  } catch {
+    return;
+  }
+
+  const jsonlFiles = await findCodexJsonlFiles(codexSessionsDir);
+
+  for (const filePath of jsonlFiles) {
+    if (getTotalMatches() >= limit || isAborted()) break;
+
+    try {
+      const fileStream = fsSync.createReadStream(filePath);
+      const rl = readline.createInterface({ input: fileStream, crlfDelay: Infinity });
+
+      // First pass: read session_meta to check project path match
+      let sessionMeta = null;
+      for await (const line of rl) {
+        if (!line.trim()) continue;
+        try {
+          const entry = JSON.parse(line);
+          if (entry.type === 'session_meta' && entry.payload) {
+            sessionMeta = entry.payload;
+            break;
+          }
+        } catch { continue; }
+      }
+
+      // Skip sessions that don't belong to this project
+      if (!sessionMeta) continue;
+      const sessionProjectPath = normalizeComparablePath(sessionMeta.cwd);
+      if (sessionProjectPath !== normalizedProjectPath) continue;
+
+      // Second pass: re-read file to find matching messages
+      const fileStream2 = fsSync.createReadStream(filePath);
+      const rl2 = readline.createInterface({ input: fileStream2, crlfDelay: Infinity });
+      let lastUserMessage = null;
+      const matches = [];
+
+      for await (const line of rl2) {
+        if (getTotalMatches() >= limit || isAborted()) break;
+        if (!line.trim()) continue;
+
+        let entry;
+        try { entry = JSON.parse(line); } catch { continue; }
+
+        let text = null;
+        let role = null;
+
+        if (entry.type === 'event_msg' && entry.payload?.type === 'user_message' && entry.payload.message) {
+          text = entry.payload.message;
+          role = 'user';
+          lastUserMessage = text;
+        } else if (entry.type === 'response_item' && entry.payload?.type === 'message') {
+          const contentParts = entry.payload.content || [];
+          if (entry.payload.role === 'user') {
+            text = contentParts
+              .filter(p => p.type === 'input_text' && p.text)
+              .map(p => p.text)
+              .join(' ');
+            role = 'user';
+            if (text) lastUserMessage = text;
+          } else if (entry.payload.role === 'assistant') {
+            text = contentParts
+              .filter(p => p.type === 'output_text' && p.text)
+              .map(p => p.text)
+              .join(' ');
+            role = 'assistant';
+          }
+        }
+
+        if (!text || !role) continue;
+        const textLower = text.toLowerCase();
+        if (!allWordsMatch(textLower)) continue;
+
+        if (matches.length < 2) {
+          const { snippet, highlights } = buildSnippet(text, textLower);
+          matches.push({ role, snippet, highlights, timestamp: entry.timestamp || null, provider: 'codex' });
+          addMatches(1);
+        }
+      }
+
+      if (matches.length > 0) {
+        projectResult.sessions.push({
+          sessionId: sessionMeta.id,
+          provider: 'codex',
+          sessionSummary: lastUserMessage
+            ? (lastUserMessage.length > 50 ? lastUserMessage.substring(0, 50) + '...' : lastUserMessage)
+            : 'Codex Session',
+          matches
+        });
+      }
+    } catch {
+      continue;
+    }
+  }
+}
+
+async function searchGeminiSessionsForProject(
+  projectPath, projectResult, words, allWordsMatch,
+  buildSnippet, limit, getTotalMatches, addMatches
+) {
+  // 1) Search in-memory sessions (created via UI)
+  for (const [sessionId, session] of sessionManager.sessions) {
+    if (getTotalMatches() >= limit) break;
+    if (session.projectPath !== projectPath) continue;
+
+    const matches = [];
+    for (const msg of session.messages) {
+      if (getTotalMatches() >= limit) break;
+      if (msg.role !== 'user' && msg.role !== 'assistant') continue;
+
+      const text = typeof msg.content === 'string' ? msg.content
+        : Array.isArray(msg.content) ? msg.content.filter(p => p.type === 'text').map(p => p.text).join(' ')
+        : '';
+      if (!text) continue;
+
+      const textLower = text.toLowerCase();
+      if (!allWordsMatch(textLower)) continue;
+
+      if (matches.length < 2) {
+        const { snippet, highlights } = buildSnippet(text, textLower);
+        matches.push({
+          role: msg.role, snippet, highlights,
+          timestamp: msg.timestamp ? msg.timestamp.toISOString() : null,
+          provider: 'gemini'
+        });
+        addMatches(1);
+      }
+    }
+
+    if (matches.length > 0) {
+      const firstUserMsg = session.messages.find(m => m.role === 'user');
+      const summary = firstUserMsg?.content
+        ? (typeof firstUserMsg.content === 'string'
+          ? (firstUserMsg.content.length > 50 ? firstUserMsg.content.substring(0, 50) + '...' : firstUserMsg.content)
+          : 'Gemini Session')
+        : 'Gemini Session';
+
+      projectResult.sessions.push({
+        sessionId,
+        provider: 'gemini',
+        sessionSummary: summary,
+        matches
+      });
+    }
+  }
+
+  // 2) Search Gemini CLI sessions on disk (~/.gemini/tmp/<project>/chats/*.json)
+  const normalizedProjectPath = normalizeComparablePath(projectPath);
+  if (!normalizedProjectPath) return;
+
+  const geminiTmpDir = path.join(os.homedir(), '.gemini', 'tmp');
+  try {
+    await fs.access(geminiTmpDir);
+  } catch {
+    return;
+  }
+
+  const trackedSessionIds = new Set();
+  for (const [sid] of sessionManager.sessions) {
+    trackedSessionIds.add(sid);
+  }
+
+  let projectDirs;
+  try {
+    projectDirs = await fs.readdir(geminiTmpDir);
+  } catch {
+    return;
+  }
+
+  for (const projectDir of projectDirs) {
+    if (getTotalMatches() >= limit) break;
+
+    const projectRootFile = path.join(geminiTmpDir, projectDir, '.project_root');
+    let projectRoot;
+    try {
+      projectRoot = (await fs.readFile(projectRootFile, 'utf8')).trim();
+    } catch {
+      continue;
+    }
+
+    if (normalizeComparablePath(projectRoot) !== normalizedProjectPath) continue;
+
+    const chatsDir = path.join(geminiTmpDir, projectDir, 'chats');
+    let chatFiles;
+    try {
+      chatFiles = await fs.readdir(chatsDir);
+    } catch {
+      continue;
+    }
+
+    for (const chatFile of chatFiles) {
+      if (getTotalMatches() >= limit) break;
+      if (!chatFile.endsWith('.json')) continue;
+
+      try {
+        const filePath = path.join(chatsDir, chatFile);
+        const data = await fs.readFile(filePath, 'utf8');
+        const session = JSON.parse(data);
+        if (!session.messages || !Array.isArray(session.messages)) continue;
+
+        const cliSessionId = session.sessionId || chatFile.replace('.json', '');
+        if (trackedSessionIds.has(cliSessionId)) continue;
+
+        const matches = [];
+        let firstUserText = null;
+
+        for (const msg of session.messages) {
+          if (getTotalMatches() >= limit) break;
+
+          const role = msg.type === 'user' ? 'user'
+            : (msg.type === 'gemini' || msg.type === 'assistant') ? 'assistant'
+            : null;
+          if (!role) continue;
+
+          let text = '';
+          if (typeof msg.content === 'string') {
+            text = msg.content;
+          } else if (Array.isArray(msg.content)) {
+            text = msg.content
+              .filter(p => p.text)
+              .map(p => p.text)
+              .join(' ');
+          }
+          if (!text) continue;
+
+          if (role === 'user' && !firstUserText) firstUserText = text;
+
+          const textLower = text.toLowerCase();
+          if (!allWordsMatch(textLower)) continue;
+
+          if (matches.length < 2) {
+            const { snippet, highlights } = buildSnippet(text, textLower);
+            matches.push({
+              role, snippet, highlights,
+              timestamp: msg.timestamp || null,
+              provider: 'gemini'
+            });
+            addMatches(1);
+          }
+        }
+
+        if (matches.length > 0) {
+          const summary = firstUserText
+            ? (firstUserText.length > 50 ? firstUserText.substring(0, 50) + '...' : firstUserText)
+            : 'Gemini CLI Session';
+
+          projectResult.sessions.push({
+            sessionId: cliSessionId,
+            provider: 'gemini',
+            sessionSummary: summary,
+            matches
+          });
+        }
+      } catch {
+        continue;
+      }
+    }
+  }
+}
+
+async function getGeminiCliSessions(projectPath) {
+  const normalizedProjectPath = normalizeComparablePath(projectPath);
+  if (!normalizedProjectPath) return [];
+
+  const geminiTmpDir = path.join(os.homedir(), '.gemini', 'tmp');
+  try {
+    await fs.access(geminiTmpDir);
+  } catch {
+    return [];
+  }
+
+  const sessions = [];
+  let projectDirs;
+  try {
+    projectDirs = await fs.readdir(geminiTmpDir);
+  } catch {
+    return [];
+  }
+
+  for (const projectDir of projectDirs) {
+    const projectRootFile = path.join(geminiTmpDir, projectDir, '.project_root');
+    let projectRoot;
+    try {
+      projectRoot = (await fs.readFile(projectRootFile, 'utf8')).trim();
+    } catch {
+      continue;
+    }
+
+    if (normalizeComparablePath(projectRoot) !== normalizedProjectPath) continue;
+
+    const chatsDir = path.join(geminiTmpDir, projectDir, 'chats');
+    let chatFiles;
+    try {
+      chatFiles = await fs.readdir(chatsDir);
+    } catch {
+      continue;
+    }
+
+    for (const chatFile of chatFiles) {
+      if (!chatFile.endsWith('.json')) continue;
+      try {
+        const filePath = path.join(chatsDir, chatFile);
+        const data = await fs.readFile(filePath, 'utf8');
+        const session = JSON.parse(data);
+        if (!session.messages || !Array.isArray(session.messages)) continue;
+
+        const sessionId = session.sessionId || chatFile.replace('.json', '');
+        const firstUserMsg = session.messages.find(m => m.type === 'user');
+        let summary = 'Gemini CLI Session';
+        if (firstUserMsg) {
+          const text = Array.isArray(firstUserMsg.content)
+            ? firstUserMsg.content.filter(p => p.text).map(p => p.text).join(' ')
+            : (typeof firstUserMsg.content === 'string' ? firstUserMsg.content : '');
+          if (text) {
+            summary = text.length > 50 ? text.substring(0, 50) + '...' : text;
+          }
+        }
+
+        sessions.push({
+          id: sessionId,
+          summary,
+          messageCount: session.messages.length,
+          lastActivity: session.lastUpdated || session.startTime || null,
+          provider: 'gemini'
+        });
+      } catch {
+        continue;
+      }
+    }
+  }
+
+  return sessions.sort((a, b) =>
+    new Date(b.lastActivity || 0) - new Date(a.lastActivity || 0)
+  );
+}
+
+async function getGeminiCliSessionMessages(sessionId) {
+  const geminiTmpDir = path.join(os.homedir(), '.gemini', 'tmp');
+  let projectDirs;
+  try {
+    projectDirs = await fs.readdir(geminiTmpDir);
+  } catch {
+    return [];
+  }
+
+  for (const projectDir of projectDirs) {
+    const chatsDir = path.join(geminiTmpDir, projectDir, 'chats');
+    let chatFiles;
+    try {
+      chatFiles = await fs.readdir(chatsDir);
+    } catch {
+      continue;
+    }
+
+    for (const chatFile of chatFiles) {
+      if (!chatFile.endsWith('.json')) continue;
+      try {
+        const filePath = path.join(chatsDir, chatFile);
+        const data = await fs.readFile(filePath, 'utf8');
+        const session = JSON.parse(data);
+        const fileSessionId = session.sessionId || chatFile.replace('.json', '');
+        if (fileSessionId !== sessionId) continue;
+
+        return (session.messages || []).map(msg => {
+          const role = msg.type === 'user' ? 'user'
+            : (msg.type === 'gemini' || msg.type === 'assistant') ? 'assistant'
+            : msg.type;
+
+          let content = '';
+          if (typeof msg.content === 'string') {
+            content = msg.content;
+          } else if (Array.isArray(msg.content)) {
+            content = msg.content.filter(p => p.text).map(p => p.text).join('\n');
+          }
+
+          return {
+            type: 'message',
+            message: { role, content },
+            timestamp: msg.timestamp || null
+          };
+        });
+      } catch {
+        continue;
+      }
+    }
+  }
+
+  return [];
+}
+
+export {
+  getProjects,
+  getSessions,
+  getSessionMessages,
+  parseJsonlSessions,
+  renameProject,
+  deleteSession,
+  isProjectEmpty,
+  deleteProject,
+  addProjectManually,
+  loadProjectConfig,
+  saveProjectConfig,
+  extractProjectDirectory,
+  clearProjectDirectoryCache,
+  getCodexSessions,
+  getCodexSessionMessages,
+  deleteCodexSession,
+  getGeminiCliSessions,
+  getGeminiCliSessionMessages,
+  searchConversations
+};