ixentbench 1.0.0__tar.gz

ixentbench-1.0.0/PKG-INFO

@@ -0,0 +1,208 @@
+Metadata-Version: 2.4
+Name: ixentbench
+Version: 1.0.0
+Summary: The AI benchmark beyond memorization — Causal spatial reasoning at 4×10⁸⁵ scale
+Author-email: iXentLabs <contact@ixentlabs.com>
+License-Expression: LicenseRef-Proprietary
+Project-URL: Homepage, https://ixentlabs.com
+Project-URL: Repository, https://github.com/ixentlabs/ixentbench
+Project-URL: Documentation, https://github.com/ixentlabs/ixentbench#readme
+Keywords: ai,benchmark,reasoning,llm,evaluation
+Classifier: Development Status :: 5 - Production/Stable
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Science/Research
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+Requires-Dist: click>=8.1
+Requires-Dist: requests>=2.31
+Requires-Dist: python-dotenv>=1.0
+Requires-Dist: google-genai>=1.0
+Requires-Dist: google-auth>=2.28
+Requires-Dist: google-auth-oauthlib>=1.2
+Provides-Extra: anthropic
+Requires-Dist: anthropic>=0.25; extra == "anthropic"
+Provides-Extra: openai
+Requires-Dist: openai>=1.30; extra == "openai"
+
+# iXentBench — The AI Benchmark Beyond Memorization
+
+Causal spatial reasoning at 4×10⁸⁵ scale, powered by **Caps i Caps** — the game that makes memorization impossible. 
+Developed by [iXentLabs](https://ixentlabs.com) (iXent Games S.L.)
+
+---
+
+## Installation
+
+```bash
+pip install ixentbench
+```
+
+---
+
+## Quick Start
+
+**1. Configure your session** at [ixentlabs.com](https://ixentlabs.com)  
+Login with Google, select benchmark type, level and AI model.  
+Download your `.env` file and add your API Key.
+
+**2. Run your session:**
+
+```bash
+ixentbench play --session YOUR_SESSION_ID
+```
+
+The visualizer opens automatically in your browser.
+
+---
+
+## Play Modes
+
+There are four ways to play iXentBench, designed for every type of participant — from AI engineers to pure mathematicians and AI enthusiasts.
+
+| Flag | A — BYOK | B — Local Model | C — Custom Agent | D — Sponsored |
+|---|---|---|---|---|
+| Uses LLM | ✅ Cloud API | ✅ Local | ❌ Pure code | ✅ iXentLabs |
+| API Key needed | ✅ Yours | ❌ No | ❌ No | ❌ No |
+| `--prompt-file` (optional) | ✅ Yes | ✅ Yes | ❌ No | ✅ Yes |
+| `--strategy-file` (optional)| ✅ Yes | ✅ Yes | ✅ Yes | ✅ Yes |
+
+---
+
+### A — BYOK (Bring Your Own Key)
+Use your own API Key from Gemini, Claude, GPT, ...  
+Your key **never leaves your machine** — it is read locally and never sent to our servers.
+
+```bash
+ixentbench play --session IQWUWP
+```
+
+`.env` file:
+```
+GOOGLE_API_KEY=your_key_here
+```
+
+---
+
+### B — Local Open Source Model
+Run any open source model (Llama, Mistral, Qwen...) locally using Ollama, LM Studio or llama.cpp.  
+No API Key required. Full privacy — no data leaves your machine.  
+⚠️ Inference speed metrics are excluded from global leaderboards (hardware-dependent).
+
+```bash
+# First start your local model, then:
+ixentbench play --session IQWUWP --local-url http://localhost:11434/v1
+```
+
+---
+
+### C — Custom Agent Script
+For engineers and researchers who want to solve iXentBench using **pure code** — no LLM required.  
+Write your own Python script using any algorithm you choose: Minimax, MCTS, A*, heuristics, neural networks trained from scratch — anything goes.
+
+Your script receives the full board state via `stdin` (JSON) and must return a move via `stdout` (JSON). It runs entirely on your machine.
+
+```bash
+ixentbench play --session IQWUWP --agent-script ./my_agent.py
+```
+
+Your script must output:
+```json
+{"command": "G4@P21(b=0)+90", "reasoning": "Explanation of your decision"}
+```
+
+> 💡 **Tip:** Even if your agent uses pure code, we strongly recommend adding a `--strategy-file` to explain your approach. Top results are presented to AI companies for talent opportunities (see Talent Hub below).
+
+---
+
+### D — iXentLabs Sponsored
+Select "Use iXentLabs credits" on the web — we provide the API Key.  
+No `.env` file needed. The cloud handles everything.
+
+```bash
+ixentbench play --session IQWUWP
+```
+
+---
+
+## Arena Mode (Multiplayer)
+
+```bash
+# Create a room (you become the host)
+ixentbench arena create --mode 4v4 --level 2
+
+# Join an existing room
+ixentbench arena join --room IXENT-4X7K
+```
+
+Supported modes: `1v1`, `2v2`, `4v4`, `human` (Human vs AI).
+
+---
+
+## Prompt Injection & Strategy Files
+
+### Prompt Injection (`--prompt-file`) — Optional
+Inject your own strategic prompt directly into the AI's System Prompt before the game starts.  
+Applies to modes **A, B and D** (LLM-based). Not applicable to mode C.
+
+```bash
+ixentbench play --session IQWUWP --prompt-file my_prompt.txt
+```
+
+> 💡 A well-crafted Prompt Injection can dramatically improve your AI's performance. Experiment freely — your injection is recorded alongside your results.
+
+### Strategy File (`--strategy-file`) — Optional
+A written description of your approach — how you designed your agent, what techniques you used, what insights guided your decisions.  
+**It is never sent to the AI.** It is stored securely and linked to your results for the Talent Hub.
+
+Applies to **all modes (A, B, C and D)**.
+
+```bash
+ixentbench play --session IQWUWP --strategy-file my_strategy.txt
+```
+
+> 💡 We strongly recommend always adding a Strategy File, especially for Custom Agents (mode C). It is your opportunity to showcase your thinking to the world's leading AI companies.
+
+---
+
+## 🏆 iXentLabs Talent Hub
+
+iXentBench is more than a benchmark — it is a **talent discovery platform**.
+
+With your explicit prior consent, iXentLabs will present the top results of each benchmark category to leading AI companies and research labs. This includes your performance metrics, your Prompt Injection (if any), and your Strategy File — giving organizations a rare window into the reasoning and engineering skills behind the results.
+
+> Your personal data (name, email, payment details) is always protected and never shared.  
+> Only anonymized benchmark data is presented, identified solely by your chosen nickname and avatar, unless you opt in to full visibility.
+
+If you achieve an exceptional result and want your work to be seen by the teams building the future of AI — **iXentBench is your stage**.
+
+---
+
+## Utility Commands
+
+```bash
+ixentbench login          # Force Google re-authentication
+ixentbench status         # Show credentials and session status
+ixentbench prompts list   # List your locally saved prompt files
+ixentbench --version      # Show version
+```
+
+---
+
+## Security
+
+- API Keys are **always read from your local `.env`** — never passed as arguments
+- Google authentication uses **OAuth 2.0** — credentials saved locally in `~/ixentbench/`
+- Game data (moves, reasoning, metrics) is stored by iXentLabs per our [Terms of Service](https://ixentlabs.com/terms)
+- Personal data (name, email) is protected per [Privacy Policy](https://ixentlabs.com/privacy)
+
+---
+
+## License
+
+Proprietary — © 2026 iXentLabs (iXent Games S.L.)  
+Contact: contact@ixentlabs.com

ixentbench-1.0.0/README.md

@@ -0,0 +1,177 @@
+# iXentBench — The AI Benchmark Beyond Memorization
+
+Causal spatial reasoning at 4×10⁸⁵ scale, powered by **Caps i Caps** — the game that makes memorization impossible. 
+Developed by [iXentLabs](https://ixentlabs.com) (iXent Games S.L.)
+
+---
+
+## Installation
+
+```bash
+pip install ixentbench
+```
+
+---
+
+## Quick Start
+
+**1. Configure your session** at [ixentlabs.com](https://ixentlabs.com)  
+Login with Google, select benchmark type, level and AI model.  
+Download your `.env` file and add your API Key.
+
+**2. Run your session:**
+
+```bash
+ixentbench play --session YOUR_SESSION_ID
+```
+
+The visualizer opens automatically in your browser.
+
+---
+
+## Play Modes
+
+There are four ways to play iXentBench, designed for every type of participant — from AI engineers to pure mathematicians and AI enthusiasts.
+
+| Flag | A — BYOK | B — Local Model | C — Custom Agent | D — Sponsored |
+|---|---|---|---|---|
+| Uses LLM | ✅ Cloud API | ✅ Local | ❌ Pure code | ✅ iXentLabs |
+| API Key needed | ✅ Yours | ❌ No | ❌ No | ❌ No |
+| `--prompt-file` (optional) | ✅ Yes | ✅ Yes | ❌ No | ✅ Yes |
+| `--strategy-file` (optional)| ✅ Yes | ✅ Yes | ✅ Yes | ✅ Yes |
+
+---
+
+### A — BYOK (Bring Your Own Key)
+Use your own API Key from Gemini, Claude, GPT, ...  
+Your key **never leaves your machine** — it is read locally and never sent to our servers.
+
+```bash
+ixentbench play --session IQWUWP
+```
+
+`.env` file:
+```
+GOOGLE_API_KEY=your_key_here
+```
+
+---
+
+### B — Local Open Source Model
+Run any open source model (Llama, Mistral, Qwen...) locally using Ollama, LM Studio or llama.cpp.  
+No API Key required. Full privacy — no data leaves your machine.  
+⚠️ Inference speed metrics are excluded from global leaderboards (hardware-dependent).
+
+```bash
+# First start your local model, then:
+ixentbench play --session IQWUWP --local-url http://localhost:11434/v1
+```
+
+---
+
+### C — Custom Agent Script
+For engineers and researchers who want to solve iXentBench using **pure code** — no LLM required.  
+Write your own Python script using any algorithm you choose: Minimax, MCTS, A*, heuristics, neural networks trained from scratch — anything goes.
+
+Your script receives the full board state via `stdin` (JSON) and must return a move via `stdout` (JSON). It runs entirely on your machine.
+
+```bash
+ixentbench play --session IQWUWP --agent-script ./my_agent.py
+```
+
+Your script must output:
+```json
+{"command": "G4@P21(b=0)+90", "reasoning": "Explanation of your decision"}
+```
+
+> 💡 **Tip:** Even if your agent uses pure code, we strongly recommend adding a `--strategy-file` to explain your approach. Top results are presented to AI companies for talent opportunities (see Talent Hub below).
+
+---
+
+### D — iXentLabs Sponsored
+Select "Use iXentLabs credits" on the web — we provide the API Key.  
+No `.env` file needed. The cloud handles everything.
+
+```bash
+ixentbench play --session IQWUWP
+```
+
+---
+
+## Arena Mode (Multiplayer)
+
+```bash
+# Create a room (you become the host)
+ixentbench arena create --mode 4v4 --level 2
+
+# Join an existing room
+ixentbench arena join --room IXENT-4X7K
+```
+
+Supported modes: `1v1`, `2v2`, `4v4`, `human` (Human vs AI).
+
+---
+
+## Prompt Injection & Strategy Files
+
+### Prompt Injection (`--prompt-file`) — Optional
+Inject your own strategic prompt directly into the AI's System Prompt before the game starts.  
+Applies to modes **A, B and D** (LLM-based). Not applicable to mode C.
+
+```bash
+ixentbench play --session IQWUWP --prompt-file my_prompt.txt
+```
+
+> 💡 A well-crafted Prompt Injection can dramatically improve your AI's performance. Experiment freely — your injection is recorded alongside your results.
+
+### Strategy File (`--strategy-file`) — Optional
+A written description of your approach — how you designed your agent, what techniques you used, what insights guided your decisions.  
+**It is never sent to the AI.** It is stored securely and linked to your results for the Talent Hub.
+
+Applies to **all modes (A, B, C and D)**.
+
+```bash
+ixentbench play --session IQWUWP --strategy-file my_strategy.txt
+```
+
+> 💡 We strongly recommend always adding a Strategy File, especially for Custom Agents (mode C). It is your opportunity to showcase your thinking to the world's leading AI companies.
+
+---
+
+## 🏆 iXentLabs Talent Hub
+
+iXentBench is more than a benchmark — it is a **talent discovery platform**.
+
+With your explicit prior consent, iXentLabs will present the top results of each benchmark category to leading AI companies and research labs. This includes your performance metrics, your Prompt Injection (if any), and your Strategy File — giving organizations a rare window into the reasoning and engineering skills behind the results.
+
+> Your personal data (name, email, payment details) is always protected and never shared.  
+> Only anonymized benchmark data is presented, identified solely by your chosen nickname and avatar, unless you opt in to full visibility.
+
+If you achieve an exceptional result and want your work to be seen by the teams building the future of AI — **iXentBench is your stage**.
+
+---
+
+## Utility Commands
+
+```bash
+ixentbench login          # Force Google re-authentication
+ixentbench status         # Show credentials and session status
+ixentbench prompts list   # List your locally saved prompt files
+ixentbench --version      # Show version
+```
+
+---
+
+## Security
+
+- API Keys are **always read from your local `.env`** — never passed as arguments
+- Google authentication uses **OAuth 2.0** — credentials saved locally in `~/ixentbench/`
+- Game data (moves, reasoning, metrics) is stored by iXentLabs per our [Terms of Service](https://ixentlabs.com/terms)
+- Personal data (name, email) is protected per [Privacy Policy](https://ixentlabs.com/privacy)
+
+---
+
+## License
+
+Proprietary — © 2026 iXentLabs (iXent Games S.L.)  
+Contact: contact@ixentlabs.com

ixentbench-1.0.0/ixentbench/__init__.py

@@ -0,0 +1,4 @@
+# -*- coding: utf-8 -*-
+__version__ = "1.0.0"
+__author__  = "iXentLabs (iXent Games SL)"
+__email__   = "antoniguasch@ixentgames.com"

ixentbench-1.0.0/ixentbench/arena.py

@@ -0,0 +1,66 @@
+# -*- coding: utf-8 -*-
+"""
+ixentbench/arena.py
+Modo Arena — esqueleto completo listo para activar
+cuando ixent-arena esté desplegado en Google Cloud Run.
+"""
+
+import click
+import requests
+from ixentbench.auth import get_firebase_token
+from ixentbench.config import IXENT_SDK_URL, validate_env
+
+
+def create_room(mode: str, level: int):
+    """Crea una sala Arena y espera a los demás jugadores."""
+    validate_env()
+    click.echo(f"\n🎮 Creating Arena room — Mode: {mode.upper()} | Level: {level}")
+
+    jwt     = get_firebase_token()
+    headers = {"X-Firebase-Token": jwt, "Content-Type": "application/json"}
+
+    resp = requests.post(f"{IXENT_SDK_URL}/arena/create", headers=headers, json={
+        "mode":  mode,
+        "level": level,
+    }, timeout=30)
+
+    if not resp.ok:
+        click.echo(f"❌ Error creating room: {resp.text}", err=True)
+        raise SystemExit(1)
+
+    data      = resp.json()
+    room_code = data.get("room_code")
+
+    click.echo(f"\n✅ Room created!")
+    click.echo(f"   Code: {room_code}")
+    click.echo(f"\n   Share this code with your opponents:")
+    click.echo(f"   ixentbench arena join --room {room_code}\n")
+    click.echo("⏳ Waiting for players to join...")
+
+    # TODO: polling hasta que todos estén listos + arranque automático
+    # Se implementa cuando ixent-arena esté desplegado en GCP
+    click.echo("\n⚠️  Arena multiplayer — coming soon!")
+
+
+def join_room(room_code: str):
+    """Se une a una sala Arena existente."""
+    validate_env()
+    click.echo(f"\n🎮 Joining Arena room: {room_code}")
+
+    jwt     = get_firebase_token()
+    headers = {"X-Firebase-Token": jwt, "Content-Type": "application/json"}
+
+    resp = requests.post(f"{IXENT_SDK_URL}/arena/join", headers=headers, json={
+        "room_code": room_code,
+    }, timeout=30)
+
+    if not resp.ok:
+        click.echo(f"❌ Error joining room: {resp.text}", err=True)
+        raise SystemExit(1)
+
+    data = resp.json()
+    click.echo(f"✅ Joined room {room_code} as {data.get('player_slot', '?')}")
+    click.echo("⏳ Waiting for host to start the game...")
+
+    # TODO: escucha de inicio + game loop Arena
+    click.echo("\n⚠️  Arena multiplayer — coming soon!")

ixentbench-1.0.0/ixentbench/auth.py

@@ -0,0 +1,85 @@
+# -*- coding: utf-8 -*-
+"""
+ixentbench/auth.py
+Google OAuth → Firebase JWT.
+Primera ejecución: abre el navegador para Sign in with Google.
+Siguientes ejecuciones: refresco silencioso desde credentials.json
+"""
+
+import requests
+import click
+from google_auth_oauthlib.flow import InstalledAppFlow
+from google.oauth2.credentials import Credentials
+from google.auth.transport.requests import Request
+from ixentbench.config import (
+    CREDENTIALS, FIREBASE_API_KEY,
+    OAUTH_CLIENT_ID, OAUTH_CLIENT_SECRET
+)
+
+SCOPES = ["openid", "https://www.googleapis.com/auth/userinfo.email"]
+
+
+def get_firebase_token() -> str:
+    """
+    Devuelve un Firebase ID Token válido (1h).
+    Gestiona automáticamente el ciclo de vida de las credenciales.
+    """
+    creds = None
+
+    if CREDENTIALS.exists():
+        try:
+            creds = Credentials.from_authorized_user_file(str(CREDENTIALS), SCOPES)
+        except Exception:
+            creds = None  # Credenciales corruptas — re-autenticar
+
+    if not creds or not creds.valid:
+        if creds and creds.expired and creds.refresh_token:
+            try:
+                creds.refresh(Request())
+            except Exception:
+                creds = None  # Token revocado — re-autenticar
+
+        if not creds:
+            flow = InstalledAppFlow.from_client_config(
+                {
+                    "installed": {
+                        "client_id":     OAUTH_CLIENT_ID,
+                        "client_secret": OAUTH_CLIENT_SECRET,
+                        "redirect_uris": ["urn:ietf:wg:oauth:2.0:oob", "http://localhost"],
+                        "auth_uri":      "https://accounts.google.com/o/oauth2/auth",
+                        "token_uri":     "https://oauth2.googleapis.com/token",
+                    }
+                },
+                SCOPES
+            )
+            creds = flow.run_local_server(port=0, open_browser=True)
+
+        CREDENTIALS.parent.mkdir(parents=True, exist_ok=True)
+        CREDENTIALS.write_text(creds.to_json())
+
+    # Intercambiar Google token → Firebase ID token
+    url = (
+        f"https://identitytoolkit.googleapis.com/v1/"
+        f"accounts:signInWithIdp?key={FIREBASE_API_KEY}"
+    )
+    resp = requests.post(url, json={
+        "postBody":            f"access_token={creds.token}&providerId=google.com",
+        "requestUri":          "http://localhost",
+        "returnSecureToken":   True,
+        "returnIdpCredential": True,
+    }, timeout=15)
+    resp.raise_for_status()
+    return resp.json().get("idToken")
+
+
+def force_login():
+    """Elimina las credenciales guardadas y fuerza re-autenticación."""
+    if CREDENTIALS.exists():
+        CREDENTIALS.unlink()
+        click.echo("🗑️  Credentials cleared.")
+    click.echo("🔐 Opening browser for Google Sign-In...")
+    token = get_firebase_token()
+    if token:
+        click.echo("✅ Login successful. Credentials saved.")
+    else:
+        click.echo("❌ Login failed.", err=True)

ixentbench-1.0.0/ixentbench/cli.py

@@ -0,0 +1,103 @@
+# -*- coding: utf-8 -*-
+"""
+ixentbench/cli.py
+Punto de entrada CLI — comando `ixentbench`
+"""
+
+import click
+from ixentbench import __version__
+
+
+@click.group()
+@click.version_option(version=__version__, prog_name="ixentbench")
+def main():
+    """
+    iXentBench — The AI benchmark beyond memorization.
+
+    Causal spatial reasoning at 4×10⁸⁵ scale.
+    Developed by iXentLabs (iXent Games S.L.)
+    """
+    pass
+
+
+# =============================================================================
+# ixentbench play
+# =============================================================================
+@main.command()
+@click.option("--session",       required=True,  help="Session ID from iXentLabs web portal")
+@click.option("--prompt-file",   default=None,   help="Prompt Injection .txt (added to SYSTEM PROMPT)")
+@click.option("--strategy-file", default=None,   help="Strategy .txt (logged for Talent Hub, NOT sent to AI)")
+@click.option("--local-url",     default=None,   help="Local model URL — e.g. http://localhost:11434/v1")
+@click.option("--agent-script",  default=None,   help="Custom agent .py script path")
+def play(session, prompt_file, strategy_file, local_url, agent_script):
+    """Run iXentBench Solo (BYOK, Local model, Custom agent or Sponsored)."""
+    from ixentbench.play import run_play
+    run_play(
+        session_id    = session,
+        prompt_file   = prompt_file,
+        strategy_file = strategy_file,
+        local_url     = local_url,
+        agent_script  = agent_script,
+    )
+
+
+# =============================================================================
+# ixentbench arena
+# =============================================================================
+@main.group()
+def arena():
+    """iXentBench Arena — Multiplayer mode (1v1, 2v2, 4v4, Human vs AI)."""
+    pass
+
+
+@arena.command("create")
+@click.option("--mode",  default="1v1",
+              type=click.Choice(["1v1", "2v2", "4v4", "human"], case_sensitive=False),
+              help="Game mode")
+@click.option("--level", default=1, type=click.IntRange(1, 4), help="Level (1-4)")
+def arena_create(mode, level):
+    """Create a new Arena room. You become the host."""
+    from ixentbench.arena import create_room
+    create_room(mode=mode, level=level)
+
+
+@arena.command("join")
+@click.option("--room", required=True, help="Room code — e.g. IXENT-4X7K")
+def arena_join(room):
+    """Join an existing Arena room."""
+    from ixentbench.arena import join_room
+    join_room(room_code=room)
+
+
+# =============================================================================
+# ixentbench login / status / prompts
+# =============================================================================
+@main.command()
+def login():
+    """Force Google re-authentication."""
+    from ixentbench.auth import force_login
+    force_login()
+
+
+@main.command()
+def status():
+    """Show current credentials and session status."""
+    from ixentbench.config import show_status
+    show_status()
+
+
+@main.group()
+def prompts():
+    """Manage your saved Prompt Injection files."""
+    pass
+
+
+@prompts.command("list")
+def prompts_list():
+    """List all saved prompt injection files."""
+    from ixentbench.config import list_prompts
+    list_prompts()
+
+
+if __name__ == "__main__":
+    main()