alvin-bot 4.15.2 → 4.16.0

package/skills/browse/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: Browser Automation
-description: 3-tier browser control — stealth scraping, CDP with persistent cookies, visual oversight. Navigate, screenshot, extract text, interact with logged-in pages.
+description: 3-tier browser control — WebFetch for plain pages, stealth scraping for JS/Cloudflare, CDP with persistent cookies for login-walled sites. Navigate, screenshot, extract text, interact with logged-in pages.
 triggers: browse, browser, test webapp, test app, test website, screenshot page, interact with, click on, fill form, visual test, qa test, check page, open page, test my app, browse to, open url, puppeteer, playwright, browser automation, linkedin, stepstone, indeed, scrape, fetch page, crawl, teste die seite, teste die app, schau dir an, öffne die seite, teste mal, visual check, check the ui, check the page, webseite öffnen, seite abrufen
 priority: 8
 category: automation
@@ -14,97 +14,112 @@ Du hast drei Browser-Strategien plus WebFetch. **Wähle die billigste passende S
 
 | Task | Tool | Warum |
 |------|------|-------|
-| Einzelne öffentliche Seite, nur Text | WebFetch oder `curl` | Am schnellsten, keine Browser-Engine |
+| Einzelne öffentliche Seite, nur Text | `curl` oder WebFetch | Am schnellsten, keine Browser-Engine |
 | Öffentliche Seite mit JS / Cloudflare | **Tier 1 Stealth** | Headless + Fingerprint-Masking |
-| Login-pflichtige Seite (LinkedIn, Gmail, …) | **Tier 2 CDP** | Echtes Chrome, persistente Cookies |
-| Komplexer Multi-Step-Flow, User soll zusehen | **Tier 3 Extension** | Visuelle Kontrolle |
+| Login-pflichtige Seite (LinkedIn, Gmail, …) | **Tier 2 CDP** | Echtes Chromium, persistente Cookies |
+| Komplexer Multi-Step-Flow, User soll zusehen | **Tier 3 Extension** | Nur in interaktiven CLI-Sessions |
 
-**NIEMALS** `scripts/browse-server.cjs` nutzen — existiert nicht mehr. **NIEMALS** nacktes `node -e "const {chromium}…"` für externe Seiten — wird sofort geblockt.
+**NIEMALS** nacktes `node -e "const {chromium}…"` für externe Seiten — wird sofort geblockt.
 
 ---
 
-## Tier 0 — WebFetch / curl (schnellster Pfad)
+## Tier 0 — curl / WebFetch (schnellster Pfad)
 
 Für statische Seiten oder APIs, die keine JS-Rendering brauchen:
 
 ```bash
-# Direkter curl
 curl -sL -H "User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36" \
-  "https://www.michaelpage.de/jobs/it-director" | htmlq -t "h1, .job-title"
-
-# Oder das WebFetch-Tool, wenn verfügbar (interpretiert Inhalt direkt)
+  "https://example.com/public-page"
 ```
 
-Wenn das einen 403/Captcha gibt → eskaliere auf Tier 1.
+Wenn das einen 403/Captcha liefert → eskaliere auf Tier 1.
 
 ---
 
 ## Tier 1 — Playwright Stealth (headless, schnell, maskiert)
 
-**Router-Script:** `~/.claude/hub/SCRIPTS/browser.sh`
+Für Seiten mit JS-Rendering oder Bot-Detection. Der Bot hat eine eingebaute Stealth-Pipeline; keine Hub-Scripts nötig.
 
-```bash
-# Seite laden, JSON-Metadata zurück (title, url, html_length)
-~/.claude/hub/SCRIPTS/browser.sh stealth "https://www.stepstone.de/jobs/it-delivery"
+**Empfohlener Weg — Bot-API:** Der interne `browser-manager` wählt automatisch die richtige Strategie. Für Scripts direkt nutzbar:
 
-# Mit Screenshot (PNG in ~/.claude/hub/BROWSER/screenshots/)
-~/.claude/hub/SCRIPTS/browser.sh stealth "https://example.com" --screenshot=page.png
+```bash
+# Falls ein externes Dev-Hub-Script vorhanden ist, kann es genutzt werden:
+# ~/.claude/hub/SCRIPTS/browser.sh stealth "<url>"
 ```
 
-**Was du bekommst:** JSON mit `{title, url, html_length, screenshot}`. Der volle HTML liegt nicht in stdout — zum Parsen den `stealth.js` direkt als Modul importieren oder `/tmp/`-File lesen.
+Ansonsten direkt über Playwright in einem kurzen Node-Script:
+
+```bash
+node -e "
+(async () => {
+  const { chromium } = require('playwright');
+  const b = await chromium.launch({ headless: true });
+  const p = await b.newPage({ userAgent: 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36' });
+  await p.goto(process.argv[1], { waitUntil: 'domcontentloaded', timeout: 30000 });
+  console.log(JSON.stringify({ url: p.url(), title: await p.title() }));
+  await b.close();
+})();
+" "https://example.com"
+```
 
-**Wann blockt das:** reCAPTCHA v3, aggressive Cloudflare, Login-Walls.
+**Wann blockt das:** reCAPTCHA v3, aggressive Cloudflare, Login-Walls → eskaliere auf Tier 2.
 
-**Konkrete funktionierende Targets (Stand 2026):**
-- StepStone (alle Job-Suchen) ✅
-- Michael Page ✅
-- Hays ✅
-- Öffentliche Blog-Posts, News-Sites ✅
-- LinkedIn (ohne Login) ❌ → Tier 2
-- Indeed / Glassdoor ❌ (403 Scraping-Block) → nur über E-Mail-Alerts
+**Konkrete funktionierende Targets (Stand 2026):** StepStone, Michael Page, Hays, Blogs, News-Sites.
 
 ---
 
-## Tier 2 — Chrome CDP (persistent Profile, echte Cookies)
+## Tier 2 — Chromium CDP (Bot-managed, persistent Profile)
 
-Echtes Chrome mit Profil unter `~/.claude/hub/BROWSER/profile/`. Login-Cookies für LinkedIn/Gmail/etc. bleiben über Sessions erhalten.
+Echtes Chromium mit Profil unter `~/.alvin-bot/browser/profile/`. Login-Cookies für LinkedIn/Gmail/etc. bleiben über Sessions erhalten.
+
+**Bot-CLI (empfohlen — funktioniert auf jedem Alvin-Bot-Install):**
 
 ```bash
-# Einmal starten (checkt ob schon läuft)
-~/.claude/hub/SCRIPTS/browser.sh cdp start headless   # headless — für Cron/Daemon
-~/.claude/hub/SCRIPTS/browser.sh cdp start headful    # sichtbar — wenn User zusehen soll
+# Starten (headless als Default — perfekt für Cron/Daemon)
+alvin-bot browser start
+alvin-bot browser start headful       # sichtbar, wenn User zusehen soll
 
 # Navigieren
-~/.claude/hub/SCRIPTS/browser.sh cdp goto "https://www.linkedin.com/jobs/search/?keywords=IT+Director"
+alvin-bot browser goto "https://www.linkedin.com/jobs/search/?keywords=IT+Director"
+
+# Screenshot → speichert nach ~/.alvin-bot/browser/screenshots/
+alvin-bot browser shot "https://www.linkedin.com/feed/" linkedin_feed.png
 
-# Screenshot
-~/.claude/hub/SCRIPTS/browser.sh cdp shot "https://www.linkedin.com/feed/" linkedin_feed.png
+# JS in Seite ausführen
+alvin-bot browser eval "https://example.com" "document.title"
 
 # Tabs auflisten
-~/.claude/hub/SCRIPTS/browser.sh cdp tabs
+alvin-bot browser tabs
+
+# Beenden (meistens nicht nötig — Chromium läuft persistent bis Bot-Neustart)
+alvin-bot browser stop
 
-# Stoppen (meistens nicht nötig, Chrome läuft persistent)
-~/.claude/hub/SCRIPTS/browser.sh cdp stop
+# Diagnose bei Problemen
+alvin-bot browser doctor
 ```
 
-**Login-Setup (einmalig):** Falls LinkedIn ausgeloggt ist, Ali per Telegram fragen:
-> "Bitte einmal in Chrome (Hub-Profil) bei LinkedIn einloggen. Cookies bleiben dann dauerhaft erhalten."
+**Architektur:** Der Bot nutzt Playwright's gebundeltes Chromium ("Google Chrome for Testing"), nicht das normale User-Chrome. Keine LaunchServices-Kollision mit parallel laufendem Chrome. Erste Einrichtung nach `npm install`:
 
-Starten mit `cdp start headful` und Chrome öffnet sichtbar → Ali loggt ein → ab dann bleiben Cookies im Profil.
+```bash
+# Playwright-Chromium einmal installieren
+npx playwright install chromium
+```
 
-**Wie teste ich ob eingeloggt:** nach `cdp goto` die URL prüfen — wenn `/authwall` oder `/login` im Pfad steht, bist du ausgeloggt.
+**Login-Setup (einmalig):** Falls die Seite ausgeloggt ist, den User fragen:
+> "Bitte einmal in Chromium (Bot-Profil) bei <Seite> einloggen. Cookies bleiben dann dauerhaft erhalten."
+
+Starten mit `alvin-bot browser start headful`, User loggt in → ab dann persistiert das Profil unter `~/.alvin-bot/browser/profile/`.
+
+**Wie teste ich ob eingeloggt:** nach `goto` die URL prüfen — wenn `/authwall` oder `/login` im Pfad steht, bist du ausgeloggt.
 
 ---
 
 ## Tier 3 — Claude-in-Chrome Extension (visuelle Kontrolle)
 
-Nur in interaktiven CLI-Sessions, nicht im Cron/Daemon.
+Nur in interaktiven Claude Code CLI-Sessions verfügbar, **nicht** im Bot-Daemon.
 
 ```bash
-# Check ob Extension verbunden
-~/.claude/hub/SCRIPTS/browser.sh ext check
-
-# Dann MCP-Tools über ToolSearch laden:
+# MCP-Tools über ToolSearch laden:
 #   mcp__claude-in-chrome__tabs_context_mcp
 #   mcp__claude-in-chrome__navigate
 #   mcp__claude-in-chrome__computer
@@ -117,45 +132,37 @@ Nur in interaktiven CLI-Sessions, nicht im Cron/Daemon.
 ## Eskalations-Regel (PFLICHT)
 
 ```
-Öffentliche Text-Seite → Tier 0 (WebFetch/curl)
+Öffentliche Text-Seite → Tier 0 (curl/WebFetch)
   ↓ 403/Cloudflare/leerer HTML?
-Tier 1 (stealth) → browser.sh stealth <url>
+Tier 1 (stealth) → Node+Playwright headless
   ↓ Captcha/Login-Wall?
-Tier 2 (CDP) → cdp start headless/headful + cdp goto <url>
+Tier 2 (CDP) → alvin-bot browser start + goto <url>
   ↓ Cookies fehlen?
-Ali fragen: "Bitte einmal in Chrome bei X einloggen, dann kann ich weitermachen."
+Den User fragen: "Bitte einmal in Chromium bei <Seite> einloggen, dann kann ich weitermachen."
 ```
 
 **NIEMALS aufgeben mit "Browser funktioniert nicht"** — es gibt immer einen nächsten Schritt. Lieber ehrlich melden "Tier 1 blockt mit Captcha, versuche Tier 2" als "Failed to load".
 
-## Status-Checks
+## Status & Diagnose
 
 ```bash
-# Übersicht aller Tiers + Health
-~/.claude/hub/SCRIPTS/browser.sh status
+# Aktueller CDP-Zustand
+alvin-bot browser status
 
-# Ist CDP Chrome gerade auf Port 9222?
+# Vollständige Diagnose (Binary, Port, PID, Profile-Lock, Chrome-Konflikt)
+alvin-bot browser doctor
+
+# Raw check ob CDP-Endpoint antwortet
 curl -s http://127.0.0.1:9222/json/version | head -c 200
 ```
 
 ## Screenshot-Ausgabe ansehen
 
-Screenshots werden gespeichert unter `~/.claude/hub/BROWSER/screenshots/` (relativ) oder dem absoluten Pfad, den du angibst. Read-Tool auf den Pfad zeigt dir das Bild direkt an.
-
-## Interaktive Ops (Klicken, Formular füllen)
-
-Für einfache Fälle: `cdp eval` mit JavaScript, das in der Seite ausgeführt wird:
-
-```bash
-~/.claude/hub/SCRIPTS/browser.sh cdp eval "https://example.com/login" \
-  "document.querySelector('#username').value='test'; document.querySelector('#password').value='pw'; document.querySelector('form').submit();"
-```
-
-Für komplexere Flows (sequentielles Klicken nach DOM-Updates) → Tier 3 (Extension) nutzen.
+Screenshots landen in `~/.alvin-bot/browser/screenshots/` (wenn nur Dateiname angegeben) oder dem absoluten Pfad. Read-Tool auf den Pfad zeigt dir das Bild direkt an.
 
 ## Wichtige Notes
 
-- **CDP-Profil-Konflikt:** Chrome kann `~/.claude/hub/BROWSER/profile/` nicht doppelt öffnen. Wenn Ali es lokal auf hatte, Port 9222 checken und `cdp stop` + `cdp start` machen.
-- **Headless vs Headful:** Im Cron/Daemon (launchd) IMMER `headless` — sonst scheitert Chrome an fehlendem Display.
-- **Nach Seiten-Navigation** (`cdp goto`) neue Tabs legt Playwright standardmäßig an — reuseTab ist nicht exponiert. Das ist OK für einzelne Scrapes, kann aber zu Tab-Explosion führen. `cdp stop` & Neustart räumt auf.
-- **Persistenz:** Cookies, LocalStorage, IndexedDB, alles in `~/.claude/hub/BROWSER/profile/`. Komplett persistiert zwischen Bot-Restarts.
+- **Profile-Konflikt:** Chromium kann `~/.alvin-bot/browser/profile/` nicht doppelt öffnen. `alvin-bot browser doctor` zeigt stale Locks.
+- **Headless vs Headful:** Im Cron/Daemon IMMER `headless` (Default) — headful scheitert an fehlendem Display.
+- **Persistenz:** Cookies, LocalStorage, IndexedDB — alles in `~/.alvin-bot/browser/profile/`. Überlebt Bot-Restarts.
+- **Kein User-Chrome-Konflikt:** Das Bot-Chromium ist ein separater Binary (Chrome-for-Testing), läuft parallel zum normalen Chrome ohne LaunchServices-Kollision.

package/skills/social-fetch/SKILL.md

@@ -78,7 +78,7 @@ print(f'Has images:  {has_images}')
 ```
 
 **Wenn yt-dlp fehlschlaegt** (private IG Posts etc.):
-1. Tier 2 CDP Browser: `~/.claude/hub/SCRIPTS/browser.sh cdp goto <url>` + Screenshot
+1. Tier 2 CDP Browser: `alvin-bot browser goto <url>` + `alvin-bot browser shot <url>`
 2. Cookies exportieren fuer yt-dlp: `--cookies /tmp/ig-cookies.txt`
 3. Oder `instaloader` fuer Instagram: `python3 -m instaloader -- -<shortcode>`
 
@@ -264,7 +264,7 @@ Nach dem Security-Check IMMER eine klare Empfehlung:
 
 ## Phase 6: Nutzwert-Bewertung
 
-### Fuer Ali/Alev-B bewerten
+### Fuer Nutzwert des Users bewerten
 
 | Dimension | Fragen |
 |-----------|--------|
@@ -382,4 +382,4 @@ Max 50 Zeichen, lowercase, Bindestriche, keine Sonderzeichen.
 - **Engagement-Metriken:** Like/View Ratio als Qualitaetssignal
 - **Creator-Profil:** Schneller Check wer der Creator ist (Expertise, Follower, Posting-Frequenz)
 - **Vergleichbare Posts:** Gibt es bessere/ausfuehrlichere Quellen zum gleichen Thema?
-- **Content-Inspiration:** Falls relevant fuer Alev-B Social — als Inspiration vormerken?
+- **Content-Inspiration:** Falls thematisch passend — als Inspiration vormerken?

package/skills/webcheck/SKILL.md

@@ -147,4 +147,4 @@ Nach dem Audit immer konkrete Empfehlungen mit Prioritaet:
 - 🟡 MITTEL — SEO/Best-Practice, bei Gelegenheit
 - 🟢 NIEDRIG — Nice-to-have
 
-Falls Hostinger hPanel noetig (CAA, DKIM, DMARC): CDP Chrome nutzen falls Session aktiv, sonst User fragen.
+Falls DNS-Provider-Login noetig (CAA, DKIM, DMARC): CDP Chrome nutzen falls Session aktiv, sonst User fragen.

package/test/async-agent-chunk-flow.test.ts

@@ -39,7 +39,7 @@ describe("async agent chunk flow (Stage 2)", () => {
       userId: 99,
       lastToolUseInput: {
         description: "SEO audit",
-        prompt: "audit gethomes.io",
+        prompt: "audit example.com",
       },
     });
 

package/test/claude-sdk-tool-use-id.test.ts

@@ -66,7 +66,7 @@ describe("claude-sdk-provider tool_use chunk contract (v4.12.1)", () => {
       input: {
         description: "full site audit",
         run_in_background: true,
-        prompt: "audit gethomes.io",
+        prompt: "audit example.com",
       },
     });
 

package/test/memory-extractor.test.ts

@@ -27,14 +27,14 @@ describe("memory-extractor (v4.11.0)", () => {
   it("parseExtractedFacts handles a clean JSON response", async () => {
     const { parseExtractedFacts } = await import("../src/services/memory-extractor.js");
     const json = JSON.stringify({
-      user_facts: ["User Ali lives in Berlin"],
+      user_facts: ["User User lives in Berlin"],
       preferences: ["Replies in German"],
-      decisions: ["Use Hostinger VPS for production"],
+      decisions: ["Use VPS VPS for production"],
     });
     const facts = parseExtractedFacts(json);
-    expect(facts.user_facts).toEqual(["User Ali lives in Berlin"]);
+    expect(facts.user_facts).toEqual(["User User lives in Berlin"]);
     expect(facts.preferences).toEqual(["Replies in German"]);
-    expect(facts.decisions).toEqual(["Use Hostinger VPS for production"]);
+    expect(facts.decisions).toEqual(["Use VPS VPS for production"]);
   });
 
   it("parseExtractedFacts handles JSON wrapped in markdown code fences", async () => {
@@ -77,33 +77,33 @@ Hope this helps!`;
   it("appendFactsToMemoryFile writes new facts under structured headers", async () => {
     const { appendFactsToMemoryFile } = await import("../src/services/memory-extractor.js");
     await appendFactsToMemoryFile({
-      user_facts: ["Ali uses launchd for the bot"],
+      user_facts: ["User uses launchd for the bot"],
       preferences: [],
       decisions: ["v4.11.0 ships memory persistence"],
     });
     const memFile = resolve(TEST_DATA_DIR, "memory", "MEMORY.md");
     expect(fs.existsSync(memFile)).toBe(true);
     const content = fs.readFileSync(memFile, "utf-8");
-    expect(content).toMatch(/Ali uses launchd for the bot/);
+    expect(content).toMatch(/User uses launchd for the bot/);
     expect(content).toMatch(/v4\.11\.0 ships memory persistence/);
   });
 
   it("appendFactsToMemoryFile dedupes on exact-string match", async () => {
     const { appendFactsToMemoryFile } = await import("../src/services/memory-extractor.js");
     await appendFactsToMemoryFile({
-      user_facts: ["Ali uses launchd for the bot"],
+      user_facts: ["User uses launchd for the bot"],
       preferences: [],
       decisions: [],
     });
     await appendFactsToMemoryFile({
-      user_facts: ["Ali uses launchd for the bot", "Ali drinks coffee"],
+      user_facts: ["User uses launchd for the bot", "User drinks coffee"],
       preferences: [],
       decisions: [],
     });
     const content = fs.readFileSync(resolve(TEST_DATA_DIR, "memory", "MEMORY.md"), "utf-8");
-    const matches = content.match(/Ali uses launchd for the bot/g);
+    const matches = content.match(/User uses launchd for the bot/g);
     expect(matches).toHaveLength(1); // not duplicated
-    expect(content).toMatch(/Ali drinks coffee/);
+    expect(content).toMatch(/User drinks coffee/);
   });
 
   it("appendFactsToMemoryFile returns 0 when all facts are duplicates", async () => {

package/test/memory-layers.test.ts

@@ -40,11 +40,11 @@ describe("memory-layers (v4.11.0)", () => {
   it("loads identity.md as L0 always", async () => {
     fs.writeFileSync(
       resolve(TEST_DATA_DIR, "memory", "identity.md"),
-      "# Identity\n\nName: Ali Levin\nLocation: Berlin",
+      "# Identity\n\nName: Test User\nLocation: Berlin",
     );
     const { loadMemoryLayers } = await import("../src/services/memory-layers.js");
     const layered = loadMemoryLayers();
-    expect(layered.identity).toMatch(/Ali Levin/);
+    expect(layered.identity).toMatch(/Test User/);
   });
 
   it("loads preferences.md as L1 always", async () => {
@@ -70,8 +70,8 @@ describe("memory-layers (v4.11.0)", () => {
   it("loads projects/*.md and exposes them with their filename as topic", async () => {
     fs.mkdirSync(resolve(TEST_DATA_DIR, "memory", "projects"), { recursive: true });
     fs.writeFileSync(
-      resolve(TEST_DATA_DIR, "memory", "projects", "alev-b.md"),
-      "# Alev-B\nVPS: 72.62.34.230, runs nginx + pm2",
+      resolve(TEST_DATA_DIR, "memory", "projects", "my-project.md"),
+      "# my-project\nVPS: 10.0.0.1, runs nginx + pm2",
     );
     fs.writeFileSync(
       resolve(TEST_DATA_DIR, "memory", "projects", "homes.md"),
@@ -81,14 +81,14 @@ describe("memory-layers (v4.11.0)", () => {
     const layered = loadMemoryLayers();
     expect(layered.projects).toHaveLength(2);
     const topics = layered.projects.map(p => p.topic).sort();
-    expect(topics).toEqual(["alev-b", "homes"]);
+    expect(topics).toEqual(["homes", "my-project"]);
     expect(layered.projects.find(p => p.topic === "homes")?.content).toMatch(/homes_production/);
   });
 
   it("buildLayeredContext returns all L0+L1 plus matching L2 by topic keyword", async () => {
     fs.writeFileSync(
       resolve(TEST_DATA_DIR, "memory", "identity.md"),
-      "Name: Ali",
+      "Name: User",
     );
     fs.writeFileSync(
       resolve(TEST_DATA_DIR, "memory", "preferences.md"),
@@ -100,24 +100,24 @@ describe("memory-layers (v4.11.0)", () => {
       "HOMES uses Postgres",
     );
     fs.writeFileSync(
-      resolve(TEST_DATA_DIR, "memory", "projects", "alev-b.md"),
-      "Alev-B uses MySQL",
+      resolve(TEST_DATA_DIR, "memory", "projects", "my-project.md"),
+      "my-project uses MySQL",
     );
 
     const { buildLayeredContext } = await import("../src/services/memory-layers.js");
 
     // Query mentions HOMES → only the homes project should be loaded
     const ctx = buildLayeredContext("Tell me about HOMES backups");
-    expect(ctx).toMatch(/Name: Ali/); // L0
+    expect(ctx).toMatch(/Name: User/); // L0
     expect(ctx).toMatch(/Be terse/); // L1
     expect(ctx).toMatch(/HOMES uses Postgres/); // L2 matched
-    expect(ctx).not.toMatch(/Alev-B uses MySQL/); // L2 not matched
+    expect(ctx).not.toMatch(/my-project uses MySQL/); // L2 not matched
   });
 
   it("buildLayeredContext without a query returns L0+L1 only (boot-up brief)", async () => {
     fs.writeFileSync(
       resolve(TEST_DATA_DIR, "memory", "identity.md"),
-      "Name: Ali",
+      "Name: User",
     );
     fs.writeFileSync(
       resolve(TEST_DATA_DIR, "memory", "preferences.md"),
@@ -131,14 +131,14 @@ describe("memory-layers (v4.11.0)", () => {
 
     const { buildLayeredContext } = await import("../src/services/memory-layers.js");
     const ctx = buildLayeredContext();
-    expect(ctx).toMatch(/Name: Ali/);
+    expect(ctx).toMatch(/Name: User/);
     expect(ctx).not.toMatch(/Postgres/); // L2 only loaded with a query
   });
 
   it("token budget: layered context truncates long projects to fit budget", async () => {
     fs.writeFileSync(
       resolve(TEST_DATA_DIR, "memory", "identity.md"),
-      "Name: Ali",
+      "Name: User",
     );
     fs.mkdirSync(resolve(TEST_DATA_DIR, "memory", "projects"), { recursive: true });
     const longContent = "homes ".repeat(2000); // ~10000 chars
@@ -155,7 +155,7 @@ describe("memory-layers (v4.11.0)", () => {
   it("monolithic MEMORY.md and split files coexist (split takes priority, mono is secondary)", async () => {
     fs.writeFileSync(
       resolve(TEST_DATA_DIR, "memory", "identity.md"),
-      "Name: Ali",
+      "Name: User",
     );
     fs.writeFileSync(
       resolve(TEST_DATA_DIR, "memory", "MEMORY.md"),
@@ -163,7 +163,7 @@ describe("memory-layers (v4.11.0)", () => {
     );
     const { buildLayeredContext } = await import("../src/services/memory-layers.js");
     const ctx = buildLayeredContext("anything");
-    expect(ctx).toMatch(/Name: Ali/); // L0
+    expect(ctx).toMatch(/Name: User/); // L0
     expect(ctx).toMatch(/Old fact still there/); // legacy still included
   });
 });

package/test/memory-sdk-injection.test.ts

@@ -20,7 +20,7 @@ beforeEach(() => {
   fs.mkdirSync(resolve(TEST_DATA_DIR, "memory"), { recursive: true });
   fs.writeFileSync(
     resolve(TEST_DATA_DIR, "memory", "MEMORY.md"),
-    "# Long-term Memory\n\n- User Ali prefers terse answers.\n- HOMES uses Postgres `homes_production`.\n",
+    "# Long-term Memory\n\n- User User prefers terse answers.\n- HOMES uses Postgres `homes_production`.\n",
   );
   process.env.ALVIN_DATA_DIR = TEST_DATA_DIR;
   vi.resetModules();
@@ -30,14 +30,14 @@ describe("SDK memory injection (v4.11.0)", () => {
   it("buildSystemPrompt(isSDK=true) now includes MEMORY.md content", async () => {
     const { buildSystemPrompt } = await import("../src/services/personality.js");
     const prompt = buildSystemPrompt(true, "en", "1234");
-    expect(prompt).toMatch(/User Ali prefers terse answers/);
+    expect(prompt).toMatch(/User User prefers terse answers/);
     expect(prompt).toMatch(/HOMES uses Postgres/);
   });
 
   it("non-SDK still gets memory injection (regression check)", async () => {
     const { buildSystemPrompt } = await import("../src/services/personality.js");
     const prompt = buildSystemPrompt(false, "en", "1234");
-    expect(prompt).toMatch(/User Ali prefers terse answers/);
+    expect(prompt).toMatch(/User User prefers terse answers/);
   });
 
   it("no MEMORY.md → SDK prompt builds without crash and without memory section", async () => {
@@ -141,6 +141,6 @@ describe("SDK smart prompt (semantic recall) on first turn (v4.11.0)", () => {
 
     const prompt = await buildSmartSystemPrompt(true, "en", "test query", "1234", true);
     expect(prompt).toBeTruthy();
-    expect(prompt).toMatch(/User Ali prefers terse answers/); // base still works
+    expect(prompt).toMatch(/User User prefers terse answers/); // base still works
   });
 });

package/test/memory-stress-restart.test.ts

@@ -164,7 +164,7 @@ describe("memory persistence stress (v4.11.0)", () => {
   it("memory-layers handles unicode in identity and projects", async () => {
     fs.writeFileSync(
       resolve(TEST_DATA_DIR, "memory", "identity.md"),
-      "Name: Ali Levin 🦊\nLocation: Berlin",
+      "Name: Test User 🦊\nLocation: Berlin",
     );
     fs.mkdirSync(resolve(TEST_DATA_DIR, "memory", "projects"), { recursive: true });
     fs.writeFileSync(
@@ -255,7 +255,7 @@ describe("memory persistence stress (v4.11.0)", () => {
   });
 
   it("layered context with very long identity stays under budget", async () => {
-    const longIdentity = "Name: Ali. ".repeat(2000); // 22000 chars
+    const longIdentity = "Name: User. ".repeat(2000); // 22000 chars
     fs.writeFileSync(
       resolve(TEST_DATA_DIR, "memory", "identity.md"),
       longIdentity,

package/test/multi-session-stress.test.ts

@@ -40,9 +40,9 @@ function writeWs(name: string, purpose: string, body: string, channels: string[]
 
 describe("multi-session stress (v4.12.0)", () => {
   it("5 parallel Slack channels each get isolated sessions", async () => {
-    writeWs("alev-b", "Alev-B dev", "Alev-B persona", ["C_ALEV"]);
+    writeWs("my-project", "my-project dev", "my-project persona", ["C_ALEV"]);
     writeWs("homes", "HOMES SaaS", "HOMES persona", ["C_HOMES"]);
-    writeWs("jobsnack", "JobSnack app", "JobSnack persona", ["C_JOBS"]);
+    writeWs("my-landing", "my-landing app", "my-landing persona", ["C_JOBS"]);
     writeWs("perseus", "Trading bot", "Perseus persona", ["C_PERSEUS"]);
     writeWs("alvin", "Bot development", "Alvin persona", ["C_ALVIN"]);
 
@@ -51,9 +51,9 @@ describe("multi-session stress (v4.12.0)", () => {
     initWorkspaces();
 
     const channels = [
-      { id: "C_ALEV", ws: "alev-b" },
+      { id: "C_ALEV", ws: "my-project" },
       { id: "C_HOMES", ws: "homes" },
-      { id: "C_JOBS", ws: "jobsnack" },
+      { id: "C_JOBS", ws: "my-landing" },
       { id: "C_PERSEUS", ws: "perseus" },
       { id: "C_ALVIN", ws: "alvin" },
     ];
@@ -81,9 +81,9 @@ describe("multi-session stress (v4.12.0)", () => {
   });
 
   it("survives full restart: 5 workspaces + 5 sessions persisted and rehydrated", async () => {
-    writeWs("alev-b", "Alev-B", "persona", ["C_ALEV"]);
+    writeWs("my-project", "my-project", "persona", ["C_ALEV"]);
     writeWs("homes", "HOMES", "persona", ["C_HOMES"]);
-    writeWs("jobsnack", "JobSnack", "persona", ["C_JOBS"]);
+    writeWs("my-landing", "my-landing", "persona", ["C_JOBS"]);
     writeWs("perseus", "Perseus", "persona", ["C_PERSEUS"]);
     writeWs("alvin", "Alvin", "persona", ["C_ALVIN"]);
 
@@ -126,13 +126,13 @@ describe("multi-session stress (v4.12.0)", () => {
     const { getSession, getCostByWorkspace } = await import("../src/services/session.js");
 
     const a = getSession("slack:C_A");
-    a.workspaceName = "alev-b";
+    a.workspaceName = "my-project";
     a.totalCost = 0.10;
     a.messageCount = 3;
     a.toolUseCount = 5;
 
     const b = getSession("slack:C_B");
-    b.workspaceName = "alev-b";
+    b.workspaceName = "my-project";
     b.totalCost = 0.05;
     b.messageCount = 2;
     b.toolUseCount = 1;
@@ -144,10 +144,10 @@ describe("multi-session stress (v4.12.0)", () => {
     c.toolUseCount = 8;
 
     const breakdown = getCostByWorkspace();
-    expect(breakdown["alev-b"].sessionCount).toBe(2);
-    expect(breakdown["alev-b"].messageCount).toBe(5);
-    expect(breakdown["alev-b"].toolUseCount).toBe(6);
-    expect(breakdown["alev-b"].totalCost).toBeCloseTo(0.15, 10);
+    expect(breakdown["my-project"].sessionCount).toBe(2);
+    expect(breakdown["my-project"].messageCount).toBe(5);
+    expect(breakdown["my-project"].toolUseCount).toBe(6);
+    expect(breakdown["my-project"].totalCost).toBeCloseTo(0.15, 10);
     expect(breakdown["homes"].sessionCount).toBe(1);
     expect(breakdown["homes"].messageCount).toBe(10);
     expect(breakdown["homes"].toolUseCount).toBe(8);
@@ -155,7 +155,7 @@ describe("multi-session stress (v4.12.0)", () => {
   });
 
   it("workspaces hot-reload picks up a new channel ID", async () => {
-    writeWs("alev-b", "Alev-B", "persona");
+    writeWs("my-project", "my-project", "persona");
     const { initWorkspaces, resolveWorkspaceOrDefault, reloadWorkspaces } =
       await import("../src/services/workspaces.js");
     initWorkspaces();
@@ -165,20 +165,20 @@ describe("multi-session stress (v4.12.0)", () => {
     expect(ws.name).toBe("default");
 
     // Add channel to config + reload
-    writeWs("alev-b", "Alev-B", "persona", ["C_NEW"]);
+    writeWs("my-project", "my-project", "persona", ["C_NEW"]);
     reloadWorkspaces();
 
     ws = resolveWorkspaceOrDefault("slack", "C_NEW", undefined);
-    expect(ws.name).toBe("alev-b");
+    expect(ws.name).toBe("my-project");
   });
 
   it("channel-name fallback finds workspace when no explicit ID mapping", async () => {
-    writeWs("alev-b", "Alev-B", "persona");
+    writeWs("my-project", "my-project", "persona");
     const { initWorkspaces, resolveWorkspaceOrDefault } = await import("../src/services/workspaces.js");
     initWorkspaces();
 
-    const ws = resolveWorkspaceOrDefault("slack", "C_UNMAPPED", "#alev-b");
-    expect(ws.name).toBe("alev-b");
+    const ws = resolveWorkspaceOrDefault("slack", "C_UNMAPPED", "#my-project");
+    expect(ws.name).toBe("my-project");
   });
 
   it("malformed workspace doesn't break loading of other workspaces", async () => {
@@ -228,7 +228,7 @@ describe("multi-session stress (v4.12.0)", () => {
   });
 
   it("simulated restart + workspace switch: workspaceName persists across flush cycles", async () => {
-    writeWs("alev-b", "Alev-B", "persona", ["C_ALEV"]);
+    writeWs("my-project", "my-project", "persona", ["C_ALEV"]);
     const { initWorkspaces, resolveWorkspaceOrDefault } = await import("../src/services/workspaces.js");
     const { buildSessionKey, getSession } = await import("../src/services/session.js");
     const { flushSessions } = await import("../src/services/session-persistence.js");
@@ -249,7 +249,7 @@ describe("multi-session stress (v4.12.0)", () => {
 
     const restored = s2.getSession("slack:C_ALEV");
     expect(restored.sessionId).toBe("alev-resume");
-    expect(restored.workspaceName).toBe("alev-b");
-    expect(restored.workingDir).toContain("alev-b");
+    expect(restored.workspaceName).toBe("my-project");
+    expect(restored.workingDir).toContain("my-project");
   });
 });

package/test/platform-session-key.test.ts

@@ -63,7 +63,7 @@ describe("buildSessionKey with string userIds (v4.12.0)", () => {
     process.env.SESSION_MODE = "per-user";
     vi.resetModules();
     const { buildSessionKey } = await import("../src/services/session.js");
-    const key = buildSessionKey("telegram", "123456", 8425689727);
-    expect(key).toBe("8425689727");
+    const key = buildSessionKey("telegram", "123456", 1234567890);
+    expect(key).toBe("1234567890");
   });
 });