feat: add cross-session recall functionality and auto-generated session titles

- Implemented `/recall <query>` command to search across all saved sessions, with options for summarizing results and resuming the top match.
- Enhanced session management by introducing AI-generated titles for sessions, improving readability in session lists and recall results.
- Updated configuration to allow users to enable or disable auto-session title generation.
- Added utility functions for managing personal configurations, including syncing personalities and custom commands across devices.
- Improved documentation to reflect new features and usage instructions.

Author: VladoIvankovic

CHANGELOG.md

@@ -11,6 +11,82 @@ For releases before v1.3.35, see [GitHub Releases](https://github.com/VladoIvank
 > as the social-share summary (IFTTT → X/Bluesky), capped at 220 chars.
 > If omitted, the feed falls back to the first paragraph.
 
+## [2.1.0] — 2026-05-19
+
+> Session memory: `/recall <query>` searches across **all** your saved sessions, `--resume` jumps straight back into the best match, `--summarize` asks the LLM what you accomplished, and sessions now get readable AI-generated titles instead of truncated first messages.
+
+### Added — `/recall` cross-session search
+
+- **`/recall <query>`** scans every saved session in the active scope
+  (project `.codeep/sessions/` when in a project, else global
+  `~/.codeep/sessions/`), matches with AND semantics (every query term
+  must appear), and ranks results by term-hit count plus a recency
+  boost. Each result shows a context snippet and the session name.
+- **`/recall <query> --resume`** loads the top-matching session
+  directly into the current conversation — skips the list + `/sessions`
+  picker dance. (TUI only; ACP shows results since it can't swap the
+  client's conversation in place.)
+- **`/recall <query> --summarize`** reads the matching sessions and
+  returns a short LLM recap of what you actually accomplished across
+  them — "ask your history a question". Works in TUI + ACP.
+- No new dependency: in-memory JSON scan, fast for the realistic
+  tens-to-hundreds-of-sessions case.
+
+### Added — portable personal config sync
+
+- **Personalities and custom commands now sync across your machines**
+  via `codeep account sync` (pull) and `codeep account push`. Global
+  ones (`~/.codeep/personalities/*.md`, `~/.codeep/commands/*.md`)
+  travel with your account alongside API keys and profiles — set up a
+  `senior-reviewer` personality or a `/deploy` command once, get it
+  everywhere. New endpoints `/api/personalities` + `/api/commands`,
+  new DB tables `user_personalities` + `user_commands`.
+- **Additive merge, never destructive**: pull only writes files that
+  don't already exist locally, so a sync can't clobber edits you
+  haven't pushed. Last-write-wins on the server via upsert.
+- **Dashboard sections** to view + delete synced personalities and
+  commands at codeep.dev/dashboard (read + prune; editing stays in the
+  CLI).
+- **Deliberately not synced**: lifecycle hooks (arbitrary shell —
+  syncing + auto-running on another machine is a security risk) and
+  MCP server configs (contain tokens). Those stay local by design.
+
+### Added — AI-generated session titles
+
+- Sessions now get a concise LLM-generated title ("OAuth2 migration
+  for auth module") instead of the first user message truncated to 60
+  chars ("help me with the…"). Generated once per session in the
+  background after it has ≥3 messages — fire-and-forget on autosave,
+  never blocks a save, never regenerates once set. Makes both
+  `/sessions` and `/recall` dramatically more readable.
+- Title priority: AI title > stored title > first-message fallback >
+  session name. Stored under `aiTitle` in the session JSON.
+- **Opt-out: `autoSessionTitle` setting** (default on). This is the
+  only feature that makes a background API call you didn't explicitly
+  request, so it's toggleable in `/settings` for privacy/cost-conscious
+  users. Off → sessions keep the first-message title, zero background
+  calls.
+
+### Changed
+
+- **`/search` description clarified** to "search the current session"
+  (vs `/recall` for cross-session) — the two were easy to confuse when
+  both said "search history".
+
+### Fixed
+
+- **`/sessions` picker showed raw session ids** (`session-2026-05-20-757cbda5`)
+  instead of readable titles. Now shows the title (AI-generated > stored
+  > first-message) with a short date + message count, so the list is
+  scannable.
+- **Models hallucinating their identity in chat mode.** Asked "which
+  model are you", GLM (and others) would claim to be Claude because the
+  chat system prompt never stated the actual identity. Both the chat
+  and agent system prompts now inject the real `model` + `provider`
+  from config, so the answer is truthful. (Agent mode already said
+  "never call yourself Claude" but didn't state the real model; now it
+  does.)
+
 ## [2.0.4] — 2026-05-19
 
 > Discoverability patch: new `/docs <command>` jumps from any slash command to its full guide on codeep.dev, the `/help` footer now points at the same place, and `/personality` and `/insights` have proper docs pages instead of one-liners.

README.md

@@ -73,7 +73,9 @@ When started in a project directory, Codeep automatically:
 - **Session picker** - Choose which session to continue on startup
 - **Per-project sessions** - Sessions stored in `.codeep/sessions/`
 - **Rename sessions** - Give meaningful names with `/rename`
-- **Search history** - Find past conversations with `/search`
+- **AI session titles** - Sessions auto-title themselves with an LLM one-liner ("OAuth2 migration for auth module") instead of a truncated first message. This makes one small background API call per session; turn it off with the `autoSessionTitle` setting (`/settings`) if you prefer zero unsolicited calls
+- **Search current session** - Find text in the open conversation with `/search`
+- **Cross-session recall** - `/recall <query>` searches across **all** saved sessions, ranked by relevance + recency. Add `--resume` to load the top match, or `--summarize` for an LLM recap of what you accomplished across matches
 - **Export** - Save to Markdown, JSON, or plain text
 - `/cost` - Per-session token usage and estimated cost (per provider/model)
 - `/compact [keepN]` - AI-summarize older messages to free up context (keeps last N, default 4)
@@ -709,19 +711,40 @@ codeep account   # Opens browser → sign in with GitHub → CLI is linked
 - **Project archiving** — hide projects from the list with one click
 - **Tasks** — create/complete bug, feature, and task items from the web or directly from the CLI with `/tasks add` and `/tasks done`
 - **API key sync** — store provider keys securely on codeep.dev, sync to any machine in one command
+- **Personal config sync** — personalities and custom slash commands sync across machines; view and prune them on the dashboard
 - **Connected devices** — see all machines linked to your account (hostname, last seen), revoke access per device
 
 ### API key sync
 
 Add keys once on the dashboard, then sync them to any machine:
 
 ```bash
-codeep account sync   # Pull keys from codeep.dev → local config
-codeep account push   # Push local keys → codeep.dev
+codeep account sync   # Pull keys + config from codeep.dev → local
+codeep account push   # Push local keys + config → codeep.dev
 ```
 
 Keys are encrypted at rest using AES-256-GCM.
 
+### Personal config sync
+
+`account sync` / `account push` also carry your **personalities** and **custom
+slash commands** between machines — the Markdown files in
+`~/.codeep/personalities/` and `~/.codeep/commands/`:
+
+```bash
+codeep account push   # Upload local personalities + commands to codeep.dev
+codeep account sync   # Download them onto a new machine
+```
+
+Merging is **additive** — `sync` only writes files that don't already exist
+locally, so it never clobbers a personality or command you've edited on this
+machine. View what's synced (and prune stale entries) under **Personalities**
+and **Custom commands** on the [dashboard](https://codeep.dev/dashboard).
+
+Hooks and MCP server configs are deliberately **not** synced: hooks run
+arbitrary shell, and MCP configs often embed tokens, so both stay local to each
+machine.
+
 ### Tasks
 
 Create, view, and complete tasks directly from the CLI — or manage them on the codeep.dev dashboard:
@@ -973,8 +996,8 @@ In `dangerous` mode, configure which tools require confirmation via `/settings`:
 | Command | Description |
 |---------|-------------|
 | `codeep account` | Link CLI to codeep.dev (GitHub OAuth) |
-| `codeep account sync` | Pull API keys from codeep.dev → local config |
-| `codeep account push` | Push local API keys → codeep.dev |
+| `codeep account sync` | Pull API keys + personalities + commands from codeep.dev → local |
+| `codeep account push` | Push local API keys + personalities + commands → codeep.dev |
 
 ### Authentication
 

src/acp/commands.ts

@@ -680,6 +680,31 @@ Anything else the agent should know — edge cases, gotchas, things to double-ch
 
     // ─── Export ────────────────────────────────────────────────────────────────
 
+    // ─── Cross-session recall (2.1.0) ─────────────────────────────────────────
+
+    case 'recall': {
+      const wantSummarize = args.includes('--summarize');
+      // --resume is TUI-only (ACP can't swap the client's conversation
+      // in place); ignore the flag here and just show results.
+      const query = args.filter(a => a !== '--resume' && a !== '--summarize').join(' ');
+      if (!query) {
+        return { handled: true, response: 'Usage: `/recall <query> [--summarize]` — searches across all saved sessions (vs `/search`, current-session only).' };
+      }
+      const { recallSessions, formatRecall, summarizeRecall } = await import('../utils/recall.js');
+      const matches = recallSessions(query, session.workspaceRoot);
+      const header = formatRecall(query, matches);
+      if (wantSummarize && matches.length > 0) {
+        onChunk('_Summarizing matching sessions…_\n\n');
+        const summary = await summarizeRecall(query, matches, session.workspaceRoot);
+        return {
+          handled: true,
+          response: summary ? `${header}\n\n---\n\n### Summary\n\n${summary}` : header,
+          streaming: true,
+        };
+      }
+      return { handled: true, response: header };
+    }
+
     // ─── Personalities + insights (2.0.3) ─────────────────────────────────────
 
     case 'personality': {

src/acp/server.ts

@@ -57,6 +57,7 @@ const AVAILABLE_COMMANDS = [
   // Sessions
   { name: 'session',   description: 'List sessions, or: new / load <name>', input: { hint: 'new | load <name>' } },
   { name: 'save',      description: 'Save current session', input: { hint: '[name]' } },
+  { name: 'recall',    description: 'Search across ALL saved sessions (cross-session)', input: { hint: '<query> [--summarize]' } },
   // Context
   { name: 'add',       description: 'Add files to agent context', input: { hint: '<file> [file2…]' } },
   { name: 'drop',      description: 'Remove files from context (no args = clear all)', input: { hint: '[file…]' } },

src/api/index.ts

@@ -271,15 +271,44 @@ const LANGUAGE_NAMES: Record<string, string> = {
   'hr': 'Croatian (Hrvatski)',
 };
 
+/**
+ * Build the agent's identity sentence from the active provider + model.
+ * Codeep is the product; the model underneath varies. Stating it
+ * explicitly stops models from guessing their own identity (GLM and
+ * others often claim to be Claude because their training data includes
+ * Claude transcripts).
+ */
+function buildIdentityLine(): string {
+  const model = String(config.get('model') || '');
+  const providerId = String(config.get('provider') || '');
+  const known: Record<string, string> = {
+    'z.ai': 'Z.AI', 'z.ai-api': 'Z.AI', 'z.ai-cn': 'Z.AI', 'z.ai-cn-api': 'Z.AI',
+    openai: 'OpenAI', anthropic: 'Anthropic', deepseek: 'DeepSeek', google: 'Google',
+    minimax: 'MiniMax', 'minimax-api': 'MiniMax', 'minimax-cn': 'MiniMax',
+    openrouter: 'OpenRouter', ollama: 'Ollama (local)',
+  };
+  const providerName = known[providerId] || providerId || 'your configured provider';
+  if (model) {
+    return `You are Codeep, an AI coding assistant. You are running on the \`${model}\` model via ${providerName}. If asked which model or provider you are, answer truthfully with these details — do not claim to be a different model.`;
+  }
+  return `You are Codeep, an AI coding assistant.`;
+}
+
 function getSystemPrompt(): string {
   const language = config.get('language');
-  
+
+  // Identity line so the model doesn't hallucinate its name when asked
+  // "what model are you" — without it, models guess from their training
+  // data (e.g. GLM claiming to be Claude). State the truth: the product
+  // is Codeep, the underlying model + provider come from config.
+  const identity = buildIdentityLine();
+
   let basePrompt: string;
   if (language === 'auto') {
-    basePrompt = `You are a helpful AI coding assistant. Always respond in the same language as the user's message. Detect the language of the user's input and reply in that same language.`;
+    basePrompt = `${identity} Always respond in the same language as the user's message. Detect the language of the user's input and reply in that same language.`;
   } else {
     const langName = LANGUAGE_NAMES[language] || 'English';
-    basePrompt = `You are a helpful AI coding assistant. Always respond in ${langName}, regardless of what language the user writes in.`;
+    basePrompt = `${identity} Always respond in ${langName}, regardless of what language the user writes in.`;
   }
 
   // Important: This is CHAT mode, not agent mode

src/config/index.ts

@@ -10,6 +10,9 @@ import { createSecureStorage, type SecureStorage } from '../utils/keychain';
 interface Session {
   name: string;
   title?: string;
+  /** LLM-generated one-liner (utils/sessionTitles.ts). Preferred over
+   *  `title` when present — see listSessionsWithInfo. */
+  aiTitle?: string;
   history: Message[];
   createdAt: string;
 }
@@ -49,6 +52,10 @@ interface ConfigSchema {
   plan: 'lite' | 'pro' | 'max';
   language: LanguageCode;
   autoSave: boolean;
+  /** Auto-generate an LLM one-liner title for sessions on save. Makes a
+   *  small background API call (uses the active model) once per session.
+   *  Default true; set false to avoid any unsolicited API calls. */
+  autoSessionTitle: boolean;
   currentSessionId: string;
   temperature: number;
   maxTokens: number;
@@ -267,6 +274,7 @@ function createConfig(): Conf<ConfigSchema> {
     plan: 'lite',
     language: 'en',
     autoSave: true,
+    autoSessionTitle: true,
     currentSessionId: '',
     temperature: 0.7,
     maxTokens: 32768,
@@ -763,23 +771,84 @@ export function saveSession(name: string, history: Message[], projectPath?: stri
     const title = firstUserMsg
       ? firstUserMsg.content.replace(/\n/g, ' ').trim().slice(0, 60)
       : name;
+    const sessionsDir = getSessionsDir(projectPath);
+    const filePath = join(sessionsDir, `${name}.json`);
+    // Preserve an existing aiTitle across re-saves so we don't regenerate.
+    let existingAiTitle: string | undefined;
+    if (existsSync(filePath)) {
+      try {
+        const prev = JSON.parse(readFileSync(filePath, 'utf-8')) as Session;
+        existingAiTitle = prev.aiTitle;
+      } catch { /* ignore corrupt prior file */ }
+    }
     const session: Session = {
       name,
       title,
+      aiTitle: existingAiTitle,
       history,
       createdAt: new Date().toISOString(),
     };
-    const sessionsDir = getSessionsDir(projectPath);
-    const filePath = join(sessionsDir, `${name}.json`);
     writeFileSync(filePath, JSON.stringify(session, null, 2));
     logSession('save', name, true);
+
+    // Fire-and-forget: generate an AI title once the session has enough
+    // content. The orchestrator early-returns if one already exists, so
+    // this is a no-op on every save after the first successful generation.
+    // Gated by autoSessionTitle so privacy/cost-conscious users can opt out
+    // of the (small, background) API call entirely.
+    if (config.get('autoSessionTitle') !== false
+        && !existingAiTitle
+        && history.filter(m => m.role !== 'system').length >= 3) {
+      void maybeGenerateSessionTitle(name, projectPath).catch(() => {});
+    }
     return true;
   } catch (error) {
     logSession('save', name, false);
     return false;
   }
 }
 
+// Guard against concurrent title generation for the same session during
+// the 5s autosave cadence — only one in-flight call per session name.
+const titlesInFlight = new Set<string>();
+
+/**
+ * Generate and persist an AI title for a session, if it doesn't have
+ * one yet. Safe to call repeatedly — early-returns when aiTitle exists
+ * or a generation is already in flight.
+ */
+export async function maybeGenerateSessionTitle(name: string, projectPath?: string): Promise<void> {
+  if (titlesInFlight.has(name)) return;
+  const sessionsDir = getSessionsDir(projectPath);
+  const filePath = join(sessionsDir, `${name}.json`);
+  if (!existsSync(filePath)) return;
+
+  let data: Session;
+  try {
+    data = JSON.parse(readFileSync(filePath, 'utf-8')) as Session;
+  } catch {
+    return;
+  }
+  if (data.aiTitle) return;
+
+  titlesInFlight.add(name);
+  try {
+    const { generateSessionTitle } = await import('../utils/sessionTitles.js');
+    const aiTitle = await generateSessionTitle(data.history);
+    if (aiTitle) {
+      // Re-read in case the session was re-saved while we were generating,
+      // so we don't clobber newer history with our stale copy.
+      try {
+        const fresh = JSON.parse(readFileSync(filePath, 'utf-8')) as Session;
+        fresh.aiTitle = aiTitle;
+        writeFileSync(filePath, JSON.stringify(fresh, null, 2));
+      } catch { /* ignore */ }
+    }
+  } finally {
+    titlesInFlight.delete(name);
+  }
+}
+
 export function loadSession(name: string, projectPath?: string): Message[] | null {
   try {
     const sessionsDir = getSessionsDir(projectPath);
@@ -902,9 +971,12 @@ export function listSessionsWithInfo(projectPath?: string): SessionInfo[] {
         const stat = statSync(filePath);
         const data = JSON.parse(readFileSync(filePath, 'utf-8')) as Session;
         const sessionName = data.name || file.replace('.json', '');
-        // Derive title: use stored title, else first user message, else session name
+        // Title priority: AI-generated one-liner > stored title > first
+        // user message > session name. aiTitle reads far better in
+        // /sessions + /recall ("OAuth2 migration" vs "help me with the…").
         const firstUserMsg = data.history?.find(m => m.role === 'user');
-        const title = data.title
+        const title = data.aiTitle
+          || data.title
           || (firstUserMsg ? firstUserMsg.content.replace(/\n/g, ' ').trim().slice(0, 60) : null)
           || sessionName;
         sessions.push({

src/renderer/App.ts

@@ -109,6 +109,7 @@ const COMMAND_DESCRIPTIONS: Record<string, string> = {
   'go': 'Execute the pending plan from /plan',
   'personality': 'Switch agent tone: concise / verbose / security / senior-reviewer / etc',
   'insights': 'Activity summary over the last N days (default 7): runs, files, tools, projects',
+  'recall': 'Search across ALL saved sessions (cross-session; /search is current-session only)',
 };
 
 import { helpCategories, keyboardShortcuts } from './components/Help';
@@ -305,6 +306,8 @@ export class App {
     'plan', 'go',
     // 2.0.3 — personalities + insights.
     'personality', 'insights',
+    // 2.1.0 — cross-session recall.
+    'recall',
     'c', 't', 'd', 'r', 'f', 'e', 'o', 'b', 'p',
   ];