Agentes gestionados en la API de Gemini: entornos aislados, habilidades y agentes persistentes
Un recorrido práctico por los Agentes Gestionados Antigravedad de Google: el entorno aislado remoto para Linux, el proxy de salida, el estado multi-turno, las habilidades, los agentes con nombre y los patrones de producción.
Fuentes: Antigravity Managed Agents Tutorial: Ship Production AI Agents
El obstáculo al que se enfrenta todo creador de agentes
Cualquiera que haya desarrollado una aplicación de IA conoce este problema. Se pide a un modelo de lenguaje que ayude a lanzar un sitio web y este responde con código ordenado. Pero luego no sucede nada, porque el modelo no puede abrir un editor, guardar el archivo ni hacer clic en “Ejecutar” para comprobar si funciona. Toda la labor pesada recae en el desarrollador.
Las herramientas existentes para agentes sí abarcan más funcionalidades, incluyendo el inicio de servidores y la ejecución de pruebas. Sin embargo, incluso al usarlas uno mismo sigue siendo necesario ocuparse de varias tareas:
- proporcionar un entorno aislado para que el modelo pueda ejecutar código sin borrar sus archivos
- escribir los mecanismos que transfieren datos entre módulos
- configurar las herramientas para que el modelo pueda llamarlas, acceder a sus datos o ejecutar un script
Siguen apareciendo más costos operativos, todos necesarios para mantener todo funcionando de manera segura. Así es como se ve cuando un modelo tiene que escribir y probar un script en Python.
from google import genai
import subprocess
client = genai.Client()
# Step 1: Ask the model for code
response = client.models.generate_content(
model="gemini-2.5-flash",
contents="Write a Python script that checks password strength and saves a report."
)
# Step 2: YOU must manually extract the code block from the text response
code = extract_code_from_markdown(response.text) # You wrote this function
# Step 3: YOU must save the file and run it locally
with open("password_checker.py", "w") as f:
f.write(code)
# ⚠️ Danger: Running untrusted AI-generated code on YOUR machine!
result = subprocess.run(["python3", "password_checker.py"], capture_output=True, text=True)
print(result.stdout)
El desarrollador se encarga de todo: extraer el código de la respuesta, las operaciones de entrada/salida de archivos, la ejecución local riesgosa y el bucle de errores que inevitablemente se activa cuando algo falla.
¿Podría una sola llamada a la API hacerse cargo de todo eso? El framework Managed Agents de Google y su agente estrella Antigravity están diseñados precisamente para ello. El modelo se convierte en un servicio: en lugar de un chatbot que solo devuelve texto, recibe una computadora en la nube segura gestionada por Google, un espacio de trabajo Linux conectado a Internet en tiempo real.
La tabla a continuación compara tres niveles de sistemas de IA.
El bucle agente principal
Ante una tarea, un agente Antigravity no se limita a adivinar. Funciona dentro de un bucle continuo respaldado por hardware real.
┌────────────────────────────────────────┐
│ 1. PLAN & REASON │
│ (Analyzes objective & breaks it down) │
└───────────────────┬────────────────────┘
│
▼
┌────────────────────────────────────────┐
│ 2. ACT │
│ (Executes a tool: Bash, Python, Web) │
└───────────────────┬────────────────────┘
│
▼
┌────────────────────────────────────────┐
│ 3. OBSERVE │
│ (Reads execution output, errors, data) │
└───────────────────┬────────────────────┘
│
└─── Loop back until task is complete
Esto va más allá de una sola solicitud y respuesta. El agente elabora un script, lo ejecuta, ve un SyntaxError en la salida de la terminal, corrige el error y vuelve a intentarlo, repitiendo el proceso hasta que la tarea se completa.
El entorno aislado remoto
Cada vez que se invoca a un agente, Google inicia un espacio de trabajo dedicado.
El resultado es una computadora Linux real en la nube, propiedad del agente durante toda la duración de su tarea.
Dos niveles detrás de la API
Trabajar con Agentes Gestionados implica dos niveles separados, a los cuales se accede a través de la API Gemini.
┌─────────────────────────────────┐
│ YOUR APP │
└────────┬───────────────┬────────┘
│ │
1. Define/Configure │ │ 2. Run Task
(System rules, ID) │ │ (Send Prompt)
▼ ▼
┌───────────────────────────────────┐ ┌───────────────────────────────────┐
│ THE CONTROL PLANE │ │ THE RUNTIME PLANE │
│ (Agents API) │ │ (Interactions API) │
├───────────────────────────────────┤ ├───────────────────────────────────┤
│ Saves persistent identity, system │ │ Spawns the Ubuntu sandbox, logs │
│ instructions, and data mounts. │ │ live traces, and processes loops. │
└───────────────────────────────────┘ └───────────────────────────────────┘
El plan de control, es decir, la API de Agents, es donde se configura y registra un agente. Es como crear un perfil de empleado: un nombre, instrucciones del sistema, habilidades personalizadas y repositorios para clonar. Devuelve un ID de Agente estático.
El plan en tiempo de ejecución, que es la API de Interacciones, es donde tiene lugar el trabajo real. Una tarea se envía a un ID de Agente como una “Interacción”, y este plano se encarga de proporcionar el entorno aislado, ejecutar herramientas, registrar los rastros del razonamiento y gestionar el ciclo de acción del agente.
Seguridad a través de un proxy de salida
Permitir que un agente autónomo ejecute scripts no verificados es arriesgado. Google isola el entorno mediante una capa de proxy de salida.
┌─────────────────────────┐
│ REMOTE UBUNTU SANDBOX │
│ (No internal secrets) │
└───────────┬─────────────┘
│ Agent attempts outbound API call
▼
┌─────────────────────────┐
│ EGRESS PROXY LAYER │ ◄── Checks Domain Allowlist
├─────────────────────────┤
│ Intercepts connection, │
│ injects secrets securely│
└───────────┬─────────────┘
│ Safe, Authenticated Request Sent
▼
[ External Target API ]
Ofrece tres garantías:
- Ningún secreto dentro de la sandbox. Las claves API y las contraseñas de la base de datos nunca se almacenan en el espacio de trabajo del agente.
- Listas de dominios permitidos. Usted decide qué dominios externos puede acceder el agente.
- Transformación de encabezados. En las llamadas a una API externa, el proxy captura la solicitud, verifica el dominio contra la lista de permitidos y agrega el token de autenticación por su cuenta. Los datos vuelven al agente mientras la clave privada permanece oculta para él.
Un primer agente en unos cinco minutos
Requisitos
- Python 3.10 o superior, siendo recomendable 3.12 ya que coincide con el entorno de ejecución de la sandbox
- google-genai SDK en versión 2.0.0 o superior (2.8.0 era la versión actual cuando se escribió esto)
Configura un entorno virtual e instala el SDK.
python3 -m venv managed-agents-env
source managed-agents-env/bin/activate # On Windows: managed-agents-env\Scripts\activate
# Install (or upgrade) the Google GenAI SDK
pip install -U google-genai
# Verify the version — must be 1.14.0 or later
python3 -c "import google.genai; print(google.genai.__version__)"
# Set your API key (get one at https://aistudio.google.com/apikey)
export GEMINI_API_KEY="your-api-key-here"
Si prefiere evitar la configuración local, Google AI Studio cuenta con una sección de plantillas “Managed Agents” que funciona como un entorno interactivo en el navegador.
Ejecutar los ejemplos
Cada ejemplo aquí es un script de Python completo y autónomo. Guárdelo en un archivo y luego ejecútelo.
# 1. Save the code to a file (an example name used here is hello_agent.py
vim hello_agent.py # or use any editor
# 2. Run it
python3 hello_agent.py
Cada interacción dura entre medio minuto y tres minutos, dependiendo de la tarea, ya que el agente inicia un entorno aislado, escribe y ejecuta código, lee los errores e itera. Las tareas sencillas se completan en menos de un minuto, mientras que las cadenas complejas de múltiples interacciones pueden tardar de dos a tres minutos por turno.
Hola, Agentes Gestionados
Compare el código torpe de varios pasos mencionado anteriormente con su equivalente en los Agentes Gestionados.
from google import genai
client = genai.Client()
# One API call. One autonomous agent. One remote Linux sandbox.
interaction = client.interactions.create(
agent="antigravity-preview-05-2026",
input="Write a Python script that generates 100 random exam scores between 0 and 100, "
"calculates the mean, median, and standard deviation, prints a summary, "
"and saves a grade distribution histogram as grade_report.png using matplotlib.",
environment="remote" # This provisions a secure Linux sandbox hosted by Google
)
print(interaction.output_text)
Ejecútalo.
python3 hello_agent.py
La primera interacción produjo el siguiente resultado. Las cifras varían de una ejecución a otra porque los datos son aleatorios.
I have successfully written and executed a Python script that generates 100 random exam scores, computes their key statistics, and plots a grade distribution histogram.
### 1. Generated Python Script (`generate_report.py`)
Here is the Python script, which is saved in the environment as `generate_report.py`. It uses `numpy` for data generation, `statistics` conventions for calculation (such as sample standard deviation), and `matplotlib` to generate a high-quality visualization.
```python
import numpy as np
import matplotlib.pyplot as plt
import statistics
def generate_and_analyze_grades():
# Set a random seed for reproducibility
np.random.seed(42)
# Generate 100 random exam scores between 0 and 100 (inclusive)
scores = np.random.randint(0, 101, size=100)
# Calculate statistics using Python's standard statistics module and NumPy
# (Using ddof=1 for sample standard deviation, which is standard for student grade samples)
mean_val = np.mean(scores)
median_val = np.median(scores)
std_val = np.std(scores, ddof=1)
# Print summary to console
print("=========================================")
print(" EXAM GRADE REPORT SUMMARY ")
print("=========================================")
print(f"Number of Students : {len(scores)}")
print(f"Minimum Score : {np.min(scores)}")
print(f"Maximum Score : {np.max(scores)}")
print(f"Mean Score : {mean_val:.2f}")
print(f"Median Score : {median_val:.2f}")
print(f"Standard Deviation : {std_val:.2f}")
print("=========================================")
# Create the histogram plot
plt.figure(figsize=(10, 6))
# Customize the grid and style
plt.grid(axis='y', linestyle='--', alpha=0.7, zorder=0)
# Draw the histogram with bins of size 10 (0-10, 10-20, ..., 90-100)
bins = range(0, 111, 10)
plt.hist(scores, bins=bins, color='#4A90E2', edgecolor='black', rwidth=0.85, zorder=3)
# Add labels and title
plt.title('Distribution of Exam Scores', fontsize=16, fontweight='bold', pad=15)
plt.xlabel('Exam Scores', fontsize=12, labelpad=10)
plt.ylabel('Number of Students', fontsize=12, labelpad=10)
# Mark Mean and Median with vertical lines
plt.axvline(mean_val, color='#D0021B', linestyle='dashed', linewidth=2, zorder=4, label=f'Mean: {mean_val:.2f}')
plt.axvline(median_val, color='#F5A623', linestyle='dotted', linewidth=2, zorder=4, label=f'Median: {median_val:.2f}')
# Set x-ticks to align with bins
plt.xticks(range(0, 101, 10))
# Add a legend
plt.legend(fontsize=11, loc='upper left')
# Add details text box on the plot
textstr = '\n'.join((
f'Total Students: {len(scores)}',
f'Mean: {mean_val:.2f}',
f'Median: {median_val:.2f}',
f'Std Dev: {std_val:.2f}'
))
props = dict(boxstyle='round', facecolor='wheat', alpha=0.5)
plt.gca().text(0.05, 0.75, textstr, transform=plt.gca().transAxes, fontsize=10,
verticalalignment='top', bbox=props)
# Adjust layout and save the figure
plt.tight_layout()
output_filename = 'grade_report.png'
plt.savefig(output_filename, dpi=300)
print(f"\nHistogram successfully saved to '{output_filename}'")
if __name__ == '__main__':
generate_and_analyze_grades()
```
---
### 2. Output Statistics Summary
Upon running the script in the sandbox environment, it produced the following output:
```text
=========================================
EXAM GRADE REPORT SUMMARY
=========================================
Number of Students : 100
Minimum Score : 1
Maximum Score : 100
Mean Score : 51.31
Median Score : 53.50
Standard Deviation : 29.70
=========================================
Histogram successfully saved to 'grade_report.png'
```
---
### 3. Generated Histogram Details (`grade_report.png`)
The script successfully outputted `grade_report.png` in high-resolution format (300 DPI) containing:
- **Clean Grid & Styling**: Standard light-gray horizontal grids to easily identify column heights.
- **Grade Bins**: Defined score ranges of size 10 (e.g., 0–10, 10–20, ..., 90–100) using custom spacing to give the columns a professional look.
- **Key Metrics Highlighted**:
- A red dashed line representing the **Mean** (`51.31`).
- A yellow dotted line representing the **Median** (`53.50`).
- **An Info Legend and Box**: In the top-left area summarizing all descriptive statistics for easy offline reference.
### How to Run it Yourself
If you want to run this script in another environment, ensure you have the required packages installed:
```bash
pip install matplotlib numpy
python generate_report.py
```
Lo que se devuelve es un resumen en inglés sencillo y no la salida del programa. Describe las acciones del agente, es decir, el código que escribió, los paquetes que instaló y los resultados obtenidos, mientras que el script y el archivo de gráfico permanecen en la caja de arena remota.
Se trata de unas pocas líneas de código comprensible. Detrás de escena, el agente:
- inició una caja de arena con Ubuntu Linux
- escribió un script en Python para generar datos aleatorios y calcular estadísticas
- instaló por su cuenta
matplotlib, sin que se le indicara que era necesario - ejecutó el script y generó un histograma en
grade_report.png - devolvió un resumen con las estadísticas y la confirmación de que el gráfico se había guardado
Una tarea más ambiciosa requiere varios herramientas juntas.
from google import genai
client = genai.Client()
interaction = client.interactions.create(
agent="antigravity-preview-05-2026",
input="""
Research the top 5 most popular Python web frameworks in 2026.
For each framework:
1. Find the latest version number and release date
2. Count the GitHub stars
3. Write a comparison table in Markdown
""",
environment="remote"
)
print(interaction.output_text)
En esta ejecución, el agente busca en Google a través de navegación web, da forma a los datos mediante ejecución de código y almacena los resultados a través de gestión de archivos, todo por sí mismo dentro de una sola llamada a la API. A continuación se muestra su salida.
The landscape of Python web development in **2026** is marked by a fascinating split [1, 2]. On one side, classic **backend-first** and **API-focused** frameworks (Django, FastAPI, Flask) remain the dominant standards for enterprise applications and microservices [3, 5]. On the other side, **pure-Python UI and full-stack** frameworks (Streamlit, Reflex) have seen explosive growth—driven by the need to build interactive web apps, data dashboards, and AI/ML interfaces without writing JavaScript or TypeScript [1, 2].
Below is an in-depth research report on the **top 5 most popular Python web frameworks in 2026**, ranked by GitHub stars and overall developer adoption.
---
### Detailed Breakdown of the Top 5 Python Web Frameworks
#### 1. FastAPI (99,034 GitHub Stars)
* **Latest Version:** `0.136.3` (Released: **May 23, 2026**)
* **Architecture Type:** ASGI (Asynchronous Server Gateway Interface)
* **Primary Focus:** High-Performance APIs and Microservices [4]
* **Core Philosophy:** FastAPI is designed to build APIs quickly with standard Python type hints. It leverages **Pydantic v2** for extremely fast data validation and serialization, and **Starlette** for high-concurrency async handling [3, 4]. It natively auto-generates interactive Swagger UI and ReDoc documentation.
* **Ideal Use Case:** AI/ML model deployment, backend APIs for single-page applications (React/Vue/Svelte), and low-latency microservices [5].
#### 2. Django (87,811 GitHub Stars)
* **Latest Version:** `6.0.6` (Released: **June 3, 2026**)
* **Architecture Type:** WSGI & ASGI (Fully async-supported)
* **Primary Focus:** "Batteries-Included" Monolithic Web Applications [4]
* **Core Philosophy:** As an 18-year industry standard, Django remains the premier full-featured web framework [2]. It handles everything—database routing (via its powerful ORM), administration panels, user authentication, form validation, and robust security defaults (SQLi, CSRF, XSS protection) [1]. In 2026, Django 6.x is highly optimized with full native asynchronous views and type hints [4].
* **Ideal Use Case:** Large-scale enterprise applications, content management systems, e-commerce platforms, and fast SaaS MVP development where you need a complete, secure architecture out of the box [2, 5].
#### 3. Flask (71,638 GitHub Stars)
* **Latest Version:** `3.1.3` (Released: **February 19, 2026**)
* **Architecture Type:** WSGI
* **Primary Focus:** Lightweight, Flexible Microframework [4]
* **Core Philosophy:** Flask provides a minimalist core, leaving the choice of database ORM, form handling, and security components completely up to the developer [1, 2]. Paired with modern 2026 extensions (SQLAlchemy 2.x, Pydantic), it represents a simple, un-opinionated foundation that never gets in the developer's way [3].
* **Ideal Use Case:** Small utility applications, lightweight APIs, microservices, and custom projects where developers want total control over their system design [1, 5].
#### 4. Streamlit (44,895 GitHub Stars)
* **Latest Version:** `1.58.0` (Released: **May 28, 2026**)
* **Architecture Type:** UI-First Scripting Paradigm
* **Primary Focus:** Rapid Data and Machine Learning Applications [1]
* **Core Philosophy:** Streamlit turns standard Python scripts into interactive, beautiful web apps in a matter of minutes. It handles the entire frontend and backend flow by re-running the script from top to bottom whenever a user interacts with a widget.
* **Ideal Use Case:** Quick data dashboards, machine learning model prototypes, and internal analytical tools for teams without frontend resources [1, 2].
#### 5. Reflex (28,467 GitHub Stars)
* **Latest Version:** `0.9.4` (Released: **June 4, 2026**)
* **Architecture Type:** Full-Stack Async React-Compiled
* **Primary Focus:** Interactive, Pure-Python Full-Stack Apps [2]
* **Core Philosophy:** Formerly known as Pynecone, Reflex compiles Python code into a high-performance **React/Next.js frontend** and a **FastAPI backend**, using real-time WebSockets to synchronize states between them [1, 2]. It bypasses JavaScript entirely, offering 60+ pre-built Radix UI components with native Tailwind integration [1, 2].
* **Ideal Use Case:** Complex, interactive web applications, real-time streaming dashboards (e.g., AI chat applications, financial trackers), and user-facing SaaS applications built entirely in Python [2].
---
### Other Honorable Mentions in 2026
While they didn't make the top 5 by GitHub stars, these frameworks are heavily utilized:
* **Tornado** (22,182 stars, v6.5.7, Released June 8, 2026): A mature, asynchronous networking framework ideal for long-lived WebSocket connections [4].
* **Sanic** (18,629 stars, v25.12.1, Released May 31, 2026): An ASGI web framework built for extreme speed and Flask-like simplicity, running on its own high-performance web server.
* **Litestar** (8,269 stars, v2.23.0, Released May 29, 2026): A highly structured, strict, and enterprise-grade ASGI alternative to FastAPI [3, 4].
---
### 2026 Python Web Frameworks Comparison Table
| Framework | GitHub Stars | Latest Version | Release Date | Architecture | Best For | Key Strength | Learning Curve |
| :--- | :--- | :--- | :--- | :--- | :--- | :--- | :--- |
| **FastAPI** | ~99,034 | `0.136.3` | May 23, 2026 | ASGI (Async) | High-performance API development & AI model deployments [4, 5] | Autogenerated Swagger docs, native Pydantic v2 validation [4] | **Easy / Moderate** |
| **Django** | ~87,811 | `6.0.6` | Jun 03, 2026 | WSGI / ASGI | Enterprise apps, content-heavy websites, secure SQL backends [1, 2] | "Batteries-included" (built-in admin, ORM, security, auth) [1, 2] | **Moderate / Hard** |
| **Flask** | ~71,638 | `3.1.3` | Feb 19, 2026 | WSGI (Sync-first) | Microservices, custom MVPs, lightweight apps [1, 5] | Unrivaled design flexibility, minimal core overhead [1] | **Very Easy** |
| **Streamlit**| ~44,895 | `1.58.0` | May 28, 2026 | Scripted Reactive | Data visualization dashboards & ML interactive tools [1] | Zero frontend experience needed; fast prototyping [1] | **Very Easy** |
| **Reflex** | ~28,467 | `0.9.4` | Jun 04, 2026 | Async React-compiled | Interactive, real-time SaaS & full-stack apps in pure Python [2] | True multi-page routing, state management without JavaScript [2] | **Moderate** |
---
### Guidance: Which Framework Should You Choose in 2026?
1. **Choose FastAPI** if you are building modern REST/GraphQL APIs, backend microservices, or deploying AI/ML models with high concurrency and automatic documentation [3, 5].
2. **Choose Django** if you are building database-driven business platforms or SaaS architectures where user authentication, administrative interfaces, and built-in security are paramount [1, 2].
3. **Choose Flask** if you need to build simple backend services or you want to hand-select every tool in your stack (e.g., combining custom database layers with external identity providers) [1].
4. **Choose Streamlit** if your core team consists of data scientists or business analysts who need to quickly present internal analytics, charts, and interactive widgets without building a complex web architecture [1, 2].
5. **Choose Reflex** if you want to build a fully interactive, production-ready full-stack application (with complex states, WebSocket interactions, and beautiful custom styles) but do not want to divide your team between Python and JavaScript [2].
---
### Sources & References
* [1] [Best Python Web Frameworks 2026 Compared - Reflex](https://reflex.dev/blog/top-python-web-frameworks/)
* [2] [Django vs Flask vs Reflex (April 2026) - Reflex](https://reflex.dev/blog/django-vs-flask-vs-reflex-comparison/)
* [3] [12 Modern Python Frameworks to Try in 2026 - Medium](https://medium.com/the-pythonworld/12-modern-python-frameworks-to-try-in-2026-e7089305bb19)
* [4] [5 top Python web frameworks of 2026 - Educative.io](https://www.educative.io/blog/top-python-web-frameworks)
* [5] [The Python Backend Framework Decision Guide for 2026 - Rollbar](https://rollbar.com/blog/python-backend-frameworks/)
Conversaciones de múltiples turnos con estado persistente
Por defecto, cada llamada a interactions.create() inicia un entorno de pruebas completamente nuevo y vacío. Para permitir que el agente continúe con el trabajo anterior, debe devolver dos IDs.
from google import genai
client = genai.Client()
# Turn 1: Research and create a report
interaction_1 = client.interactions.create(
agent="antigravity-preview-05-2026",
input="Research the current state of carbon capture technology in 2026. "
"Write a comprehensive 500-word report and save it as report.md",
environment="remote"
)
print(f"Environment ID: {interaction_1.environment_id}")
print(f"Interaction ID: {interaction_1.id}")
print(interaction_1.output_text)
# Turn 2: Build on the previous work - same sandbox, same files
interaction_2 = client.interactions.create(
agent="antigravity-preview-05-2026",
environment=interaction_1.environment_id, # ← Re-attaches to same Ubuntu sandbox
previous_interaction_id=interaction_1.id, # ← Preserves conversation memory
input="Convert that report.md file into a clean index.html webpage "
"with modern inline CSS styling and responsive design."
" Please output the contents of the index.html file"
)
print(interaction_2.output_text)
La primera interacción generó un archivo report.md.
Environment ID: 5eb9b4b8-c02f-4a8f-b7a0-c90a796c01a6
Interaction ID: v1_ChdidXNuYXJLQkF1eWlqdU1QODRmUjJRZxIXYnVzbmFyS0JBdXlpanVNUDg0ZlIyUWc
I have researched the state of carbon capture, utilization, and storage (CCUS)
technology in 2026 and compiled a comprehensive 494-word report.
The report has been successfully saved to your environment as **`report.md`**.
Below is the full text of the report,
highlighting how the CCUS sector has transitioned from pilot-stage
demonstrations to industrial infrastructure in 2026.
***
# The State of Carbon Capture Technology in 2026: From Pilot to Infrastructure
## Executive Summary
....
## Point-Source Capture: The Current Backbone
...
## Scaling Direct Air Capture (DAC)
...
## Infrastructure, Transport, and Policy
...
## Key Challenges
...
***
### Sources Cited
* **[1] S&P Global:** [2026 CCUS: Navigating the tides of the great realignment](https://www.spglobal.com/energy/en/news-research/blog/energy-transition/041426-2026-ccus-navigating-the-tides-of-the-great-realignment) (April 2026)
* **[2] Carbon Herald:** [What's Next For Carbon Capture, Utilization & Storage (CCUS) In 2026](https://carbonherald.com/whats-next-for-carbon-capture-utilization-storage-ccus-in-2026/) (January 2026)
* **[3] Nature Tech Memos:** [Top 10 Carbon Capture Startups for Corporate Partnerships in 2026](https://www.naturetechmemos.com/p/top-10-carbon-capture-startups-for-corporate-partnerships-in-2026) (April 2026)
.....
La segunda interacción reutilizó el entorno e interacción anteriores y solicitó una versión HTML del informe, index.html, con la instrucción de solicitar explícitamente el contenido HTML en la respuesta.
I have successfully converted your `report.md` file into a clean, modern, and highly responsive **`index.html`** webpage!
### What was done:
1. **Designed a cohesive palette:** Implemented a modern climate-tech inspired aesthetic using Slate (`#1e293b`) and a range of deep and light teals (`#0f766e`, `#14b8a6`, `#115e59`).
2. **Added an Interactive Stat Dashboard:** Created a high-level statistical summary card grid displaying key 2026 data at a glance (e.g., global Mtpa capacity, Stratos targets, US Section 45Q tax credits).
3. **Structured visual highlights:** Embedded two distinct grid comparison cards for the **Climeworks Mammoth** and **1PointFive Stratos** facilities, as well as a custom callout block highlighting the trans-European **Northern Lights** shipping route.
4. **Professionalized the typography & chemistry:** Used system-ui fonts for excellent loading speeds and correctly formatted all chemical formulas to standard subscripts (e.g., **CO<sub>2</sub>**).
5. **Citations & Interactivity:** Mapped the original references into clickable, superscript citation numbers (e.g., `[1]`) that link smoothly down to a beautifully bulleted "Sources Cited" section with custom index bullet styling.
The full HTML contents of the newly created `index.html` are shown below:
```html
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>The State of Carbon Capture Technology in 2026</title>
<style>
:root {
--primary: #0f766e;
--primary-light: #14b8a6;
--primary-dark: #115e59;
--bg: #f8fafc;
--card-bg: #ffffff;
--text-main: #1e293b;
--text-muted: #475569;
--accent: #0284c7;
--border: #e2e8f0;
--shadow: 0 4px 6px -1px rgba(15, 118, 110, 0.05), 0 2px 4px -2px rgba(15, 118, 110, 0.05);
--shadow-md: 0 10px 15px -3px rgba(15, 118, 110, 0.1), 0 4px 6px -4px rgba(15, 118, 110, 0.1);
}
* {
box-sizing: border-box;
}
body {
font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, Helvetica, Arial, sans-serif;
background-color: var(--bg);
color: var(--text-main);
line-height: 1.8;
margin: 0;
padding: 0;
-webkit-font-smoothing: antialiased;
}
<!-- LOTS MORE CSS STUFF :-) -->
footer {
text-align: center;
color: var(--text-muted);
font-size: 0.875rem;
margin-top: 4rem;
padding-top: 2rem;
border-top: 1px solid var(--border);
}
</style>
</head>
<body>
<header>
<div class="header-content">
<span class="meta">Special Intelligence Report</span>
<h1>The State of Carbon Capture Technology in 2026</h1>
<p>From Pilot to Infrastructure</p>
</div>
</header>
<main>
<!-- Key Metrics Dashboard -->
<div class="stats-grid">
<div class="stat-card">
<div class="stat-num">~73 Mtpa</div>
<div class="stat-label">Global Operational Capacity</div>
</div>
<div class="stat-card">
<div class="stat-num">500k Tons</div>
<div class="stat-label">Stratos DAC Annual Target</div>
</div>
<div class="stat-card">
<div class="stat-num">Up to $180</div>
<div class="stat-label">US 45Q Subsidy per Ton</div>
</div>
<div class="stat-card">
<div class="stat-num">36,000 T</div>
<div class="stat-label">Mammoth Iceland Capacity</div>
</div>
</div>
<div class="content-wrapper">
<!-- Executive Summary -->
<section id="executive-summary">
<h2>Executive Summary</h2>
<p>
In 2026, the Carbon Capture, Utilization, and Storage (CCUS) industry is undergoing an "industrial hardening" phase, transitioning decisively from demonstration-stage pilots to commercial-scale infrastructure<sup><a href="#ref-1" class="citation-link">[1]</a></sup>. Global operational capture capacity has reached approximately 73 million metric tons per annum (Mtpa) as of mid-2026, up from 50 Mtpa in early 2025<sup><a href="#ref-1" class="citation-link">[1]</a></sup><sup><a href="#ref-2" class="citation-link">[2]</a></sup>. This momentum is propelled by robust climate policies, corporate carbon-removal commitments, and the commissioning of megaton-scale facilities.
</p>
</section>
<!-- Point-Source Capture -->
<section id="point-source">
<h2>Point-Source Capture: The Current Backbone</h2>
<p>
Point-source carbon capture remains the commercial backbone of CCUS<sup><a href="#ref-2" class="citation-link">[2]</a></sup>. Technologies deployed at industrial and energy facilities—such as cement, steel, chemicals, and refining—now represent the vast majority of active capture capacity. Post-combustion chemical absorption using amine-based solvents is the most widely deployed technology<sup><a href="#ref-2" class="citation-link">[2]</a></sup>. Companies like Carbon Upcycling Technologies are successfully integrating capture systems with utilization, converting captured carbon dioxide (CO<sub>2</sub>) into high-quality construction materials, turning emissions from cement manufacturers into a low-carbon concrete feedstock<sup><a href="#ref-3" class="citation-link">[3]</a></sup>.
</p>
</section>
<!-- Scaling Direct Air Capture -->
<section id="dac">
<h2>Scaling Direct Air Capture (DAC)</h2>
<p>
Direct Air Capture is experiencing a dramatic scale-up<sup><a href="#ref-4" class="citation-link">[4]</a></sup>. Climeworks' "Mammoth" facility in Iceland, operational since May 2024, captures up to 36,000 tons of CO<sub>2</sub> annually, storing it permanently underground via Carbfix mineralization<sup><a href="#ref-4" class="citation-link">[4]</a></sup><sup><a href="#ref-5" class="citation-link">[5]</a></sup>. Meanwhile, 1PointFive's (a subsidiary of Occidental Petroleum) "Stratos" facility in Ector County, Texas, is entering active operation in 2026<sup><a href="#ref-4" class="citation-link">[4]</a></sup>. Designed to capture up to 500,000 tons of atmospheric CO<sub>2</sub> annually using liquid solvent technology licensed from Carbon Engineering, Stratos is currently the world’s largest DAC plant<sup><a href="#ref-4" class="citation-link">[4]</a></sup><sup><a href="#ref-6" class="citation-link">[6]</a></sup>.
</p>
<div class="highlight-grid">
<div class="highlight-box">
<h4>Climeworks: Mammoth (Iceland)</h4>
<p>Nameplate capacity of 36,000 tons/year. Relies on solid-sorbent collectors powered by clean geothermal energy with deep Carbfix basaltic storage.</p>
</div>
<div class="highlight-box">
<h4>1PointFive: Stratos (Texas)</h4>
<p>World's largest facility with 500,000 tons/year target. Employs liquid-solvent infrastructure designed for rapid regional scalability.</p>
</div>
</div>
<p>
Major technology firms (including Microsoft, Google, Meta, and Amazon) have signed multi-year offtake agreements for high-quality, durable carbon credits, paying between $200 and $300 per ton, though current baseline DAC capture costs remain high, between $400 and $1,000 per ton<sup><a href="#ref-3" class="citation-link">[3]</a></sup><sup><a href="#ref-7" class="citation-link">[7]</a></sup>.
</p>
</section>
<!-- Infrastructure, Transport, and Policy -->
<section id="infrastructure">
<h2>Infrastructure, Transport, and Policy</h2>
<p>
The commercial viability of carbon capture relies heavily on dedicated transportation and storage networks<sup><a href="#ref-2" class="citation-link">[2]</a></sup>. In Europe, 2026 marks the active operation of Norway’s "Northern Lights" project, the world's first open-source CO<sub>2</sub> transport and storage network<sup><a href="#ref-8" class="citation-link">[8]</a></sup>. For example, Yara's flagship Sluiskil ammonia plant in the Netherlands is liquefying up to 800,000 tons of CO<sub>2</sub> annually to be shipped by Northern Lights for permanent undersea storage<sup><a href="#ref-8" class="citation-link">[8]</a></sup>. Concurrently, Denmark’s "Greensand" offshore storage initiative is beginning operations<sup><a href="#ref-2" class="citation-link">[2]</a></sup>.
</p>
<div class="callout">
<div class="callout-title">The Trans-European Shipping Pathway</div>
<p>Northern Lights bridges emission-heavy inland industrial sites like Yara Sluiskil in the Netherlands directly with permanent injection storage wells beneath the North Sea seabed.</p>
</div>
<p>
On the policy front, the sector is heavily anchored by government subsidies. In the United States, the Inflation Reduction Act’s (IRA) modified Section 45Q tax credit provides up to $180 per metric ton for DAC and $85 per metric ton for point-source capture<sup><a href="#ref-9" class="citation-link">[9]</a></sup>. In the European Union, the Net Zero Industry Act has accelerated cross-border transport approvals, providing regulatory certainty.
</p>
</section>
<!-- Key Challenges -->
<section id="challenges">
<h2>Key Challenges</h2>
<p>
Despite rapid progress, critical bottlenecks persist. Chief among these is the high energy intensity of DAC, which requires 1.5 to 2.5 megawatt-hours (MWh) of zero-carbon energy per ton of captured CO<sub>2</sub><sup><a href="#ref-7" class="citation-link">[7]</a></sup>. Permitting delays for geologic injection wells—specifically the EPA’s rigorous Class VI permits in the United States—also restrict how fast captured carbon can be sequestered<sup><a href="#ref-4" class="citation-link">[4]</a></sup>. Overcoming these economic and infrastructure hurdles remains essential to achieving megaton targets by 2030.
</p>
</section>
<!-- Sources Cited -->
<div class="sources-section">
<h3>Sources Cited</h3>
<ul class="sources-list">
<li id="ref-1" data-index="1"><strong>S&P Global:</strong> <a href="https://www.spglobal.com/energy/en/news-research/blog/energy-transition/041426-2026-ccus-navigating-the-tides-of-the-great-realignment" target="_blank">2026 CCUS: Navigating the tides of the great realignment</a> (April 2026)</li>
<li id="ref-2" data-index="2"><strong>Carbon Herald:</strong> <a href="https://carbonherald.com/whats-next-for-carbon-capture-utilization-storage-ccus-in-2026/" target="_blank">What's Next For Carbon Capture, Utilization & Storage (CCUS) In 2026</a> (January 2026)</li>
<li id="ref-3" data-index="3"><strong>Nature Tech Memos:</strong> <a href="https://www.naturetechmemos.com/p/top-10-carbon-capture-startups-for-corporate-partnerships-in-2026" target="_blank">Top 10 Carbon Capture Startups for Corporate Partnerships in 2026</a> (April 2026)</li>
<li id="ref-4" data-index="4"><strong>Senken:</strong> <a href="https://www.senken.io/blog/top-direct-air-capture-carbon-removal-projects-buyers-guide" target="_blank">The Top 3 Direct Air Capture Carbon Removal Projects</a> (February 2026)</li>
<li id="ref-5" data-index="5"><strong>Climeworks:</strong> <a href="https://climeworks.com/plant-mammoth" target="_blank">Mammoth: our newest direct air capture and storage facility</a> (May 2024)</li>
<li id="ref-6" data-index="6"><strong>Carbon Credits:</strong> <a href="https://carboncredits.com/top-3-carbon-capture-leaders-to-drive-the-net-zero-race-in-2026/" target="_blank">Top 3 Carbon Capture Leaders to Drive the Net-Zero Race in 2026</a> (January 2026)</li>
<li id="ref-7" data-index="7"><strong>Energy Solutions Intelligence:</strong> <a href="https://energy-solutions.co/articles/sub/carbon-capture-direct-air-dac-cost-analysis" target="_blank">Direct Air Capture in 2026: Cost, Scale, and Path to $200/tCO2</a> (January 2026)</li>
<li id="ref-8" data-index="8"><strong>World Economic Forum:</strong> <a href="https://www.weforum.org/stories/2026/01/scale-carbon-capture-storage-climate-action/" target="_blank">How to scale carbon capture and storage for climate action</a> (January 2026)</li>
<li id="ref-9" data-index="9"><strong>International Energy Agency (IEA):</strong> <a href="https://www.iea.org/policies/16255-inflation-reduction-act-2022-sec-13104-extension-and-modification-of-credit-for-carbon-oxide-sequestration" target="_blank">Inflation Reduction Act 2022: Sec. 13104 Extension and Modification of Credit for Carbon Oxide Sequestration</a> (February 2026)</li>
</ul>
</div>
<footer>
<p style="text-align: center; color: var(--text-muted); font-size: 0.875rem; margin: 0;">© 2026 Carbon Capture Intelligence. Compiled June 2026.</p>
</footer>
</div>
</main>
</body>
</html>
```
El informe generado se ve así.
Los dos IDs
Los dos IDs tienen funciones diferentes, como lo resume la tabla.
Un consejo: se puede pasar environment_id sin previous_interaction_id. Esto reutiliza los archivos y paquetes instalados, pero inicia una conversación nueva, lo cual es adecuado para dividir el espacio de trabajo para una tarea nueva.
Descarga de archivos
Pedir al agente que incluya el contenido del archivo en su respuesta es una opción, aunque generalmente no es la preferida. A menudo, el archivo debe provenir directamente del entorno aislado. La lista a continuación repite la sesión anterior de varias interacciones en la que se creó un informe en formato Markdown y una versión HTML, y agrega al final código que descarga toda la instantánea del entorno como un archivo tar y la extrae en una carpeta local. Eso es lo que el entorno de Managed Agents admite actualmente.
import os
import requests
import tarfile
from google import genai
client = genai.Client()
# Turn 1: Research and create a report
interaction_1 = client.interactions.create(
agent="antigravity-preview-05-2026",
input="Research the current state of carbon capture technology in 2026. "
"Write a comprehensive 500-word report and save it as report.md",
environment="remote"
)
print(f"Environment ID: {interaction_1.environment_id}")
print(f"Interaction ID: {interaction_1.id}")
print(interaction_1.output_text)
# Turn 2: Build on the previous work - same sandbox, same files
interaction_2 = client.interactions.create(
agent="antigravity-preview-05-2026",
environment=interaction_1.environment_id, # ← Re-attaches to same Ubuntu sandbox
previous_interaction_id=interaction_1.id, # ← Preserves conversation memory
input="Convert that report.md file into a clean index.html webpage "
"with modern inline CSS styling and responsive design."
" Please output the contents of the index.html file"
)
print(interaction_2.output_text)
env_id = interaction_2.environment_id
api_key = os.environ.get("GEMINI_API_KEY")
response = requests.get(
f"https://generativelanguage.googleapis.com/v1beta/files/environment-{env_id}:download",
params={"alt": "media"},
headers={"x-goog-api-key": api_key},
allow_redirects=True,
)
with open("snapshot_env.tar", "wb") as f:
f.write(response.content)
os.makedirs("extracted_env_snapshot", exist_ok=True)
with tarfile.open("snapshot_env.tar") as tar:
tar.extractall(path="extracted_env_snapshot")
En el momento de escribir esto, no existe un método SDK para ello, por lo que la captura de estado proviene de una llamada REST directa. Después de ejecutar el programa, los archivos extraídos aparecen en una carpeta local.
Personalización de agentes con habilidades y personajes
En lugar de una única cadena de instrucciones extensa, Antigravity utiliza un sistema de configuración basado en el sistema de archivos. Cuando se inicia el entorno aislado, busca una carpeta oculta .agents/.
📁 Your-Project-Directory/
└── 📁 .agents/
├── 📄 AGENTS.md ← Global system instructions (persona, rules, standards)
└── 📁 skills/
└── 📁 data-cleaner/
└── 📄 SKILL.md ← Specific skill with step-by-step instructions
AGENTS.md como la persona
Este archivo contiene las reglas básicas, los límites de comportamiento y el contexto del proyecto para el agente, al igual que un README que se lee antes de cada tarea.
SKILL.md para conocimientos modulares
Las habilidades agrupan los conocimientos especializados en unidades reutilizables, con YAML frontmatter para metadatos y Markdown para las instrucciones.
---
name: data-cleaner
description: Use when the user needs to clean, normalize, or validate tabular data files.
---
# Data Cleaner
## When to Use
- User provides CSV, Excel, or JSON files that need cleaning
- Data has missing values, inconsistent formatting, or duplicate rows
## Steps
1. Load the data file using pandas
2. Profile the data: count nulls, duplicates, and type mismatches
3. Apply cleaning rules (fill nulls, normalize strings, deduplicate)
4. Save the cleaned output and generate a summary report
La idea central es la divulgación progresiva. El agente carga todo el contenido de un SKILL.md solo cuando decide que esa habilidad es importante para la tarea actual. Al iniciarse, lee únicamente el name y el description del frontmatter, lo que mantiene la ventana de contexto limpia.
Si una habilidad no se activa cuando se espera, la causa suele ser su description del frontmatter. Hágala más específica respecto a cuándo debe activarse dicha habilidad.
Crear un agente persistente
Para entornos de producción, se puede registrar una vez la configuración de un agente persistente para no tener que repetirla. La API de Agents crea un agente con nombre, en este caso utilizando la habilidad de limpieza de datos del apartado anterior.
from google import genai
client = genai.Client()
# Register a reusable, named agent with the data-cleaner skill baked in
agent = client.agents.create(
id="my-csv-cleaner",
base_agent="antigravity-preview-05-2026",
system_instruction="You are a data quality engineer. Always use pandas for data manipulation. "
"Always generate a before/after summary showing what changed.",
base_environment={
"type": "remote",
"sources": [
{
"type": "inline",
"target": ".agents/AGENTS.md",
"content": (
"# Data Quality Agent\n\n"
"## Standards\n"
"- Never drop rows silently — log every removal with a reason\n"
"- Normalize all string columns to lowercase, stripped of whitespace\n"
"- Output cleaned files in UTF-8 CSV format\n"
"- Always print a summary table at the end\n"
)
},
{
"type": "inline",
"target": ".agents/skills/data-cleaner/SKILL.md",
"content": (
"---\n"
"name: data-cleaner\n"
"description: Use when the user needs to clean, normalize, or validate tabular data files.\n"
"---\n"
"# Data Cleaner\n\n"
"## When to Use\n"
"- User provides CSV, Excel, or JSON files that need cleaning\n"
"- Data has missing values, inconsistent formatting, or duplicate rows\n\n"
"## Steps\n"
"1. Load the data file using pandas\n"
"2. Profile the data: count nulls, duplicates, and type mismatches\n"
"3. Apply cleaning rules (fill nulls, normalize strings, deduplicate)\n"
"4. Save the cleaned output and generate a summary report\n"
)
}
]
}
)
print(f"Agent created with ID: {agent.id}")
Al guardarlo como create_csv_cleaner_agent.py y ejecutarlo, se registra un agente persistente al que se puede acceder por su ID, my-csv-cleaner.
Para hacerlo funcionar, primero cree un archivo CSV con registros de clientes intencionadamente desordenados llamado messy_customers.csv.
name,email,phone,country
John Smith,john@example.com,+14155551234,US
jane doe, JANE@EXAMPLE.COM ,4155559999,
John Smith,john@example.com,+14155551234,US
Bob Wilson,bob@@invalid,+442071234567,
María García,maria@test.com,+34612345678,Spain
alice brown ,alice@example.com,,US
Charlie Lee,charlie@test.com,+61412345678,AU
jane doe,jane@example.com,+14155559999,
Luego, el agente puede ser invocado por su ID desde cualquier lugar de la aplicación, utilizando un script como run_csv_cleaner_agent.py.
from google import genai
client = genai.Client()
# Read your local CSV file
with open("messy_customers.csv", "r") as f:
csv_content = f.read()
# Inject the file into the sandbox and tell the agent to clean it
interaction = client.interactions.create(
agent="my-csv-cleaner",
input="Clean the file at data/customers.csv. "
"Remove duplicate rows by email (case-insensitive), "
"standardize phone numbers to E.164 format, "
"fill missing 'country' fields by looking up the phone country code, "
"and flag any rows where the email format is invalid. "
"Generate a before/after summary showing what changed.",
environment={
"type": "remote",
"sources": [
{
"type": "inline",
"target": "data/customers.csv", # ← Path inside the sandbox
"content": csv_content # ← Your local file contents
}
]
}
)
print(interaction.output_text)
El array sources de ese script carga los archivos en el entorno aislado antes de que inicie el agente. Para el agente, data/customers.csv es simplemente un archivo normal en su sistema de archivos Linux, y no tiene idea de que el archivo fue enviado desde una máquina local. Cualquier tipo de archivo funciona, incluyendo CSV, JSON, XML, scripts en Python y archivos de configuración.
Obsérvese que la instrucción no menciona nada sobre pandas, UTF-8 ni convenciones de registro. Esos detalles ya se encuentran en AGENTS.md y SKILL.md, que vienen junto con el agente especificado, y la habilidad data-cleaner se activa automáticamente porque el texto de la tarea coincide con su description. A continuación se muestra un ejemplo de ejecución.
I have successfully cleaned the customer dataset at `data/customers.csv` using Python and `pandas`.
Below is a detailed report and before/after summary of the transformations applied.
---
### Summary of Transformations
1. **whitespace Trimming:** Stripped all leading and trailing whitespace from both column names and text values (e.g., ` alice brown ` became `alice brown`).
2. **Email Normalization & Validation:**
- Converted all email addresses to lowercase and trimmed whitespace.
- Added a new column `email_invalid` (boolean) to flag emails with an invalid format.
- Identified `bob@@invalid` as an invalid format (due to a duplicate `@` symbol and lack of a top-level domain).
3. **Phone Number Standardization:**
- Standardized all valid phone numbers to the international **E.164** format using the `phonenumbers` library.
- Converted `4155559999` (with a default US region fallback) to `+14155559999`.
- Ignored the missing phone number for `alice brown` while leaving it empty.
4. **Country Code Lookup:**
- Looked up and populated missing `country` values based on the telephone country code prefix.
- Filled `jane doe`'s country as `US` (derived from prefix `+1`).
- Filled `Bob Wilson`'s country as `GB` (derived from prefix `+44`).
- Kept existing valid country entries (e.g., `Spain`, `US`, `AU`) intact.
5. **Deduplication:**
- Removed duplicate rows by performing a case-insensitive match on the cleaned `email` column, keeping the first occurrence.
- Safely removed the duplicate row for `john@example.com` (Row 3).
- Safely removed the duplicate row for `jane@example.com` (Row 8).
---
### Before/After Summary
#### 1. Data Quality Metrics
| Metric | Before Clean | After Clean | Change Description |
| :--- | :---: | :---: | :--- |
| **Total Rows** | 8 | 6 | 2 duplicate rows were removed. |
| **Duplicates Removed** | 0 | 2 | Dropped rows 3 (`john@example.com`) and 8 (`jane@example.com`). |
| **Invalid Emails Flagged** | 0 | 1 | Flagged `bob@@invalid` as invalid (`email_invalid = True`). |
| **Standardized Phone Numbers** | 1 | 5 | All numbers formatted to E.164 (e.g., `4155559999` -> `+14155559999`). |
| **Filled Country Fields** | 5 | 6 | Filled 2 missing country values (`US` and `GB`) via phone prefix lookups. |
#### 2. Row-by-Row Comparison
Below is the row-by-row evolution from the raw file to the cleaned output:
| Row # | Name (Before) | Name (After) | Email (Before) | Email (After) | Phone (Before) | Phone (After) | Country (Before) | Country (After) | Email Invalid Flag | Action Taken / Status |
| :---: | :--- | :--- | :--- | :--- | :--- | :--- | :--- | :--- | :---: | :--- |
| **1** | `John Smith` | `John Smith` | `john@example.com` | `john@example.com` | `+14155551234` | `+14155551234` | `US` | `US` | `False` | Retained as first unique instance. |
| **2** | `jane doe` | `jane doe` | ` JANE@EXAMPLE.COM ` | `jane@example.com` | `4155559999` | `+14155559999` | *Missing* | `US` | `False` | Email normalized; Phone formatted to E.164; Country populated from `+1` prefix. |
| **3** | `John Smith` | — | `john@example.com` | — | `+14155551234` | — | `US` | — | — | **Removed** (Duplicate email). |
| **4** | `Bob Wilson` | `Bob Wilson` | `bob@@invalid` | `bob@@invalid` | `+442071234567` | `+442071234567` | *Missing* | `GB` | `True` | Country populated from `+44` prefix; Email flagged as invalid. |
| **5** | `María García` | `María García` | `maria@test.com` | `maria@test.com` | `+34612345678` | `+34612345678` | `Spain` | `Spain` | `False` | Retained; Email valid; Preserved existing country. |
| **6** | ` alice brown ` | `alice brown` | `alice@example.com` | `alice@example.com` | *Missing* | *Missing* | `US` | `US` | `False` | Name whitespace stripped; Phone left empty; Preserved existing country. |
| **7** | `Charlie Lee` | `Charlie Lee` | `charlie@test.com` | `charlie@test.com` | `+61412345678` | `+61412345678` | `AU` | `AU` | `False` | Retained; Email valid. |
| **8** | `jane doe` | — | `jane@example.com` | — | `+14155559999` | — | *Missing* | — | — | **Removed** (Duplicate email). |
---
### Cleaned Dataset Output
The cleaned data has been written back to `data/customers.csv` and contains the following records:
```csv
name,email,phone,country,email_invalid
John Smith,john@example.com,+14155551234,US,False
jane doe,jane@example.com,+14155559999,US,False
Bob Wilson,bob@@invalid,+442071234567,GB,True
María García,maria@test.com,+34612345678,Spain,False
alice brown,alice@example.com,,US,False
Charlie Lee,charlie@test.com,+61412345678,AU,False
```
Casos de uso en el mundo real
Una vez establecidas las bases, algunos agentes muestran cómo se combinan las diferentes partes, y cada uno introduce una nueva capacidad.
Caso de uso uno: refactorización de código y corrección de pruebas
Tomemos un proyecto en Python con código obsoleto y pruebas fallidas. Un LLM ordinario solo proporciona fragmentos de código, dejando al desarrollador la tarea de copiarlos, ejecutar las pruebas, depurar y iterar manualmente. Con Antigravity, el agente clona el código, ejecuta el conjunto de pruebas, lee los errores del terminal, corrige el código fuente y vuelve a ejecutar las pruebas, en un ciclo hasta que todas pasen.
Un repositorio de ejemplo con código intencionalmente defectuoso sirve como objetivo. Contiene tres errores de causa raíz que, juntos, provocan seis fallos en las pruebas:
calculate_total()enutils.pyrealiza operaciones aritméticas sobre una cadenatax_ratesin convertirla a número flotante.
normalize_name() en utils.py se cae cuando se le pasa None, ya que no hay verificación de valores nulos.Invoice.summary() en models.py hace referencia a self.total, el objeto del método, en lugar de llamar a self.total().El agente que lo maneja está definido de esta manera.
from google import genai
client = genai.Client()
system_instructions = """
You are an expert QA and Refactoring Engineer. Your workflow is:
1. Clone the target repository into the workspace
2. Install all dependencies from requirements.txt
3. Run the full pytest suite and capture all output
4. For each failing test:
a. Read the error traceback carefully
b. Open the source file causing the failure
c. Diagnose the root cause
d. Apply the minimal fix
5. Re-run pytest after each fix to confirm resolution
6. Continue until ALL tests pass (0 failures)
7. Generate a summary of every change you made
IMPORTANT: Make the minimal possible change to fix each test.
Do not refactor working code unnecessarily.
"""
interaction = client.interactions.create(
agent="antigravity-preview-05-2026",
system_instruction=system_instructions,
input="""
Clone this repository: https://github.com/rominirani/uc1-legacy-app
Run the pytest suite to check for environment or code failures.
Isolate the bugs causing test failures and fix them one by one.
Keep running the tests until you achieve a 100% pass rate.
When done, provide:
1. A list of all files modified with a diff summary
2. The final pytest output showing all tests passing
3. Any dependency changes you had to make
""",
environment="remote"
)
print(interaction.output_text)
Gúrralo como uc1_refactor.py y ejecútelo.
python3 uc1_refactor.py
A continuación se muestra la salida esperada.
I have successfully diagnosed and resolved all the issues causing test failures in the `uc1-legacy-app` repository. Below is a detailed summary of the findings, modifications, and the final test execution output.
---
### 1. Modified Files & Diff Summary
#### File: `src/models.py`
- **Issue**: In the `Invoice.summary()` method, the `self.total` method itself was passed to `format_currency()` as an object, instead of being invoked (`self.total()`). This led to a `TypeError` during format evaluation.
- **Fix**: Changed `self.total` to `self.total()`.
- **Diff Summary**:
```diff
@@ -23,7 +23,7 @@ class Invoice:
"""Return a one-line summary of the invoice."""
from src.utils import format_currency, normalize_name
name = normalize_name(self.customer_name)
- return f"Invoice {self.invoice_id}: {name} - {format_currency(self.total)}"
+ return f"Invoice {self.invoice_id}: {name} - {format_currency(self.total())}"
```
#### File: `src/utils.py`
- **Issues**:
1. **`calculate_total()`**: When `tax_rate` is received as a string, Python fails to perform multiplication of float/string (`TypeError: can't multiply sequence by non-int of type 'float'`).
2. **`normalize_name()`**: When passed a `None` value (common in CSV imports), the code raised `AttributeError: 'NoneType' object has no attribute 'strip'`.
- **Fixes**:
1. Explicitly cast `subtotal` and `tax_rate` to `float` prior to calculating the tax.
2. Add a safeguard to return an empty string (`""`) when `name` is `None`.
- **Diff Summary**:
```diff
@@ -11,8 +11,10 @@ def calculate_total(subtotal, tax_rate):
Returns:
The total as a float rounded to 2 decimal places.
"""
- tax = subtotal * tax_rate / 100
- return round(subtotal + tax, 2)
+ subtotal_val = float(subtotal)
+ tax_rate_val = float(tax_rate)
+ tax = subtotal_val * tax_rate_val / 100
+ return round(subtotal_val + tax, 2)
def format_currency(amount):
@@ -36,6 +38,8 @@ def normalize_name(name):
Returns:
Cleaned, title-cased name.
"""
+ if name is None:
+ return ""
return name.strip().title()
```
---
### 2. Dependency Changes
No additional packages or dependency changes were made. All required libraries from `requirements.txt` (consisting of `pytest>=7.0.0`) were successfully installed and utilized without modification.
---
### 3. Final Pytest Output
Running the full suite confirms that all 10 tests now pass:
```
============================= test session starts ==============================
platform linux -- Python 3.12.3, pytest-9.0.3, pluggy-1.6.0
rootdir: /uc1-legacy-app
plugins: anyio-4.13.0
collected 10 items
tests/test_models.py .... [ 40% ]
tests/test_utils.py ...... [100% ]
============================== 10 passed in 0.02s ==============================
```
Sin ninguna ayuda directa, el agente:
- ejecuta
git cloneen la terminal de la sandbox - ejecuta
pip install -r requirements.txten el Python de la sandbox - ejecuta
pytest -vy lee toda la salida, incluidos los mensajes de error - abre los archivos fuente con errores y localiza la lógica defectuosa
- escribe parches, guarda los archivos y vuelve a ejecutar
pytest - repite el proceso hasta que la salida muestre cero errores
Este es el ciclo de planificar, actuar, observar en acción. El agente no solo genera código; lo ejecuta, lee la salida real de la terminal y se corrige a sí mismo.
Caso de uso dos: análisis competitivo
Digamos que un equipo necesita un análisis competitivo exhaustivo de los productos rivales. Recopilar información a mano es tedioso, y los herramientas automatizadas se dañan constantemente. En este caso, el agente busca en Google con sus herramientas de navegación, recorre los sitios de la competencia, se adapta a diferentes diseños de página, escribe su propio script en pandas para organizar los datos y exporta una hoja de cálculo finalizada.
from google import genai
client = genai.Client()
system_instructions = """
You are a precise Business Intelligence Analyst. Your workflow:
1. Use web search to find current, real-time information
2. Navigate to official product/pricing pages for each competitor
3. Extract: pricing tiers, key features, target audience, notable limitations
4. Structure ALL findings into a pandas DataFrame
5. Export to both CSV and a formatted Markdown comparison table
6. Include the date/time of research and source URLs for every data point
Rules:
- Only report data you can verify from official sources
- If pricing isn't publicly available, note "Contact Sales" - don't guess
- Use USD for all pricing normalization
"""
interaction = client.interactions.create(
agent="antigravity-preview-05-2026",
system_instruction=system_instructions,
input="""
Research the pricing models and core feature matrices of the top 3 alternative
platforms to Vercel for frontend deployment:
1. Netlify
2. Cloudflare Pages
3. Cloud Run
For each, extract:
- Free tier limits (bandwidth, builds, sites)
- Pro/paid tier pricing and what it unlocks
- Key differentiating features
- Notable limitations or complaints from developer communities
Create:
1. A competitive_matrix.csv spreadsheet with all data
2. A competitive_analysis.md report with a formatted comparison table
and a "Recommendation" section at the bottom
""",
environment="remote"
)
print(interaction.output_text)
Genera este informe.
As a Business Intelligence Analyst, I have conducted a precise competitive analysis of the top three frontend deployment alternatives to Vercel: **Netlify**, **Cloudflare Pages**, and **Google Cloud Run**.
All findings have been structured using `pandas` and exported to `/` (the current working directory) as:
1. `competitive_matrix.csv` — The raw, structured dataset.
2. `competitive_analysis.md` — A comprehensive, reader-ready report including a detailed markdown comparison matrix, architectural breakdowns, limitation analyses, and a tailored recommendation framework.
All data has been verified against official vendor pricing, product, and developer documentation as of **June 9, 2026**. All currency values are normalized in **USD**.
---
### Core Finding Highlights
#### 1. Netlify
* **The Architectural Shift (April 2026):** Netlify made a massive strategic update to its Pro plans, moving from a seat-based model ($19/member/month) to a **flat $20/month per organization** [10]. This includes unlimited team members (Owners, Developers, Reviewers, and Git Contributors) [10], which is highly disruptive compared to Vercel's strict per-seat billing.
* **The Credits Trap:** Usage is now managed entirely via a unified credit-based model (300 credits/mo free, 3,000 on Pro) [2, 10]. However, credits are consumed fast: bandwidth costs 20 credits per GB [2, 10] (meaning Free tier is only **~15 GB max** and Pro is only **~150 GB max** if solely used for traffic). If Free tier credits run out, **all site traffic pauses immediately** (no auto-recharge is supported on Free) [2, 11].
* **Capabilities:** Highly integrated features like Netlify Database (managed Postgres via Neon) [4], Blob storage, Forms, and Auth [10]. Serverless timeout is 10s on Free/Personal [2] and 26s on Pro [2].
#### 2. Cloudflare Pages
* **Unmetered Freedom:** Genuinely **unlimited and unmetered static bandwidth and requests** across all tiers (including the $0 Free tier) [24, 25, 27].
* **The Pro Tier ($20/mo billed annually / $25/mo monthly):** Tied directly to Cloudflare's Workers Paid plan [25, 27]. It increases build limits from 500/mo to 5,000/mo, concurrent builds from 1 to 5, and the file-count ceiling per project from 20,000 to 100,000 assets [31] (unlocked via the `PAGES_WRANGLER_MAJOR_VERSION=4` env variable) [31].
* **Edge Isolates Constraints:** While running serverless code on Cloudflare's global network of 310+ cities is blazingly fast with zero cold starts [24, 33], it uses a **strict V8 edge runtime** [24]. Standard Node.js library modules (`fs`, `path`, native `crypto`) do not run natively [24], forcing developers to compile meta-frameworks like Next.js via edge adapters (e.g., `@cloudflare/next-on-pages` or OpenNext) [24] which restricts certain features like Incremental Static Regeneration (ISR).
#### 3. Google Cloud Run
* **Pure Container Portability:** Cloud Run runs any application, language, or compiled binary (Node.js, Python, Go, Rust) packaged in a standard Docker container [37, 39]. It runs full Node.js natively [39], ensuring 100% Next.js feature parity (ISR, Server Actions, dynamic image optimization) with zero edge-runtime workarounds [39].
* **No Monthly Base Fee:** Operates on a pure pay-as-you-go serverless model (billed in granular vCPU-seconds and GiB-seconds) [41], backed by a generous Always Free tier (2 million requests/mo, 180,000 vCPU-seconds, 360,000 GiB-seconds, and 120 daily Cloud Build minutes free) [37, 38, 41].
* **The Infrastructure Overhead:** Not a standard "push-to-deploy" platform. Developers must write Dockerfiles, configure Artifact Registry, and manage IAM and networking [37, 39]. It is regionally hosted; to get a global CDN with SSL mapping, you must set up Cloud CDN and a Global HTTP Load Balancer [37] (~$20-$30/mo base fee) and navigate container cold starts (2 to 10+ seconds) [37] and expensive internet egress rates ($0.085 - $0.12 per GB) [37].
---
### Structure of Generated Files
#### 1. `competitive_matrix.csv`
A raw, tabular CSV structured for business intelligence pipelines, featuring the following schema:
- `Platform Name`
- `Free Tier Pricing (USD)`
- `Free Tier Bandwidth Limit`
- `Free Tier Build Limit`
- `Free Tier Sites/Projects Limit`
- `Free Tier Key Specs & Features`
- `Paid/Pro Tier Price (USD)`
- `Paid/Pro Tier Features Unlocked`
- `Key Differentiating Features`
- `Notable Limitations & Developer Complaints`
- `Date of Research`
- `Source URLs`
#### 2. `competitive_analysis.md`
A professionally formatted report including:
* **Executive Summary:** Contextualizing why frontend teams are exploring alternatives to Vercel in 2026.
* **Structured Comparison Table:** Transposed for high-density reading and side-by-side comparison.
* **Deep-Dive Section:** Dedicated analyses of the Free limits, paid features, unique differentiators, and community complaints for Netlify, Cloudflare Pages, and Google Cloud Run.
* **Architectural Recommendation Flowchart & Matrix:** Tailored decision pathways based on project profile:
* *Static / JAMstack / Zero-Budget* $\rightarrow$ **Cloudflare Pages**
* *Collaborative Teams / Rapid Prototyping / Seat-Free DX* $\rightarrow$ **Netlify**
* *Complex Server-Side Rendering / Backend Pipelines / Custom Docker* $\rightarrow$ **Google Cloud Run**
---
### Source Verification & Citations
1. **Netlify:**
- [Official Pricing & Plans](https://www.netlify.com/pricing/)
- [Introducing Seat-Free Collaborative Billing (April 14, 2026)](https://www.netlify.com/blog/pricing-netlify-for-3-billion-builders/)
- [Credit-Based Pricing Documentation](https://docs.netlify.com/manage/accounts-and-billing/billing/billing-for-credit-based-plans/credit-based-pricing-plans/)
- [Netlify Free Tier Credit Analysis (Temps)](https://temps.sh/compare/vs-netlify)
2. **Cloudflare Pages:**
- [Workers & Pages Pricing Matrix](https://www.cloudflare.com/plans/developer-platform/)
- [Official Pages Limits Documentation](https://developers.cloudflare.com/pages/platform/limits/)
- [Cloudflare Pages Platform Features](https://pages.cloudflare.com/)
- [Cloudflare Pages Edge Limitations (Temps)](https://temps.sh/compare/vs-cloudflare-pages)
3. **Google Cloud Run:**
- [Official Cloud Run Pricing Breakdown](https://cloud.google.com/run/pricing)
- [Google Cloud Free Tier Inclusions](https://cloud.google.com/free)
- [Cloud Run Quotas & Limits](https://docs.cloud.google.com/run/quotas)
- [Cloud Run Cost Optimization & Egress Guide (Cloudchipr)](https://cloudchipr.com/blog/cloud-run-pricing)
The generated files are saved directly in your working directory and are immediately available for download or integration into your reporting pipelines. Let me know if you would like me to modify the analysis parameters or explore a specific platform further!
Una continuación puede reutilizar el entorno e interacción anteriores, tomando el archivo competitive_matrix.csv generado y convirtiéndolo en una visualización. El código a continuación obtiene únicamente la respuesta de texto, pero la técnica de descarga mostrada anteriormente puede extraer los archivos generados del entorno aislado.
# Follow-up: generate a visual chart from the data we just collected
interaction_2 = client.interactions.create(
agent="antigravity-preview-05-2026",
environment=interaction.environment_id,
previous_interaction_id=interaction.id,
input="""
Using the competitive_matrix.csv you just created:
1. Create a grouped bar chart comparing the free tier limits across all 3 platforms
2. Create a pricing comparison chart for the paid tiers
3. Save both charts as PNG files with clear labels and a professional color scheme
4. Add the charts as embedded images in the competitive_analysis.md report
"""
)
print(interaction_2.output_text)
El concepto aquí es navegación por la web combinada con ejecución de código en una sola tarea. El agente busca en Internet en tiempo real y luego escribe código en Python para estructurar y visualizar lo que encontró, todo en un mismo flujo.
Patrones que funcionan en producción
Esté preparado para reconstruirlos. El rendimiento de los modelos mejora rápidamente, por lo que diseñe los agentes asumiendo que pronto se reconstruirán con modelos más recientes. Mantenga la configuración modular, ya que los archivos AGENTS.md y SKILL.md son fáciles de intercambiar, y no sobrediseñe la capa de orquestación.
Divida los agentes grandes en partes más pequeñas. Evite tener un agente masivo con una instrucción de sistema de 2,000 palabras. Descomponga los problemas complejos en subagentes especializados.
Trate los errores como entradas, no como fallos graves. En diseños de agentes sólidos, los errores son datos que el agente utiliza para reflexionar y corregirse. Lee un TypeError, analiza su causa y la soluciona. Evite los bloques try/except genéricos que ocultan los fallos y permita que el agente los vea.
Utilice evaluaciones en lugar de pruebas unitarias. El comportamiento de los agentes no es determinista, por lo que las pruebas deben centrarse en métricas de evaluación y no en la coincidencia exacta de salidas.
# ❌ Don't: Assert exact output
assert agent_output == "The answer is 42"
# ✅ Do: Evaluate behavioral success rate
results = [run_agent(task) for _ in range(20)]
success_rate = sum(1 for r in results if r.meets_criteria) / len(results)
assert success_rate >= 0.85 # Agent succeeds at least 85% of the time
Costo del token de visualización. Dado que el agente ejecuta un bucle de razonamiento autónomo, una sola solicitud puede desencadenar muchas operaciones internas, por lo que el costo por tarea varía considerablemente. Esta tabla ofrece una idea aproximada.
Utilice tiempos de espera estrictos y supervise la longitud de los registros durante el desarrollo, además de emplear system_instruction para pedirle al agente que sea conciso cuando no se requiera un razonamiento detallado.
Planifique para imprevistos. Aquí se resumen los patrones de manejo de errores para casos comunes de fallo.
Cómo funciona en el fondo
Varias mecánicas explican el comportamiento observado anteriormente:
- Una llamada a la API,
interactions.create(), inicia un bucle de orquestación del lado del servidor. Un modelo Gemini Flash continúa razonando, seleccionando herramientas, ejecutando código y revisando resultados hasta que la tarea se completa.
environment_id devuelto en llamadas posteriores, se conservan los archivos y el estado entre turnos, lo que evita crear un entorno nuevo cada vez.client.agents.create() y ser invocado desde cualquier lugar mediante su ID.Ideas para el próximo agente
- Un agente CI/CD que monitorea un repositorio de GitHub, ejecuta pruebas con cada push y abre solicitudes de corrección para los fallos
- Un generador de documentación que lee un código fuente y escribe documentación completa de la API
- Un orquestador de pipelines de datos que se conecta a un almacén, ejecuta transformaciones y valida los resultados
- Un agente de triaje de soporte que lee los tickets, evalúa su gravedad, redacta respuestas y eleva los problemas críticos
- Un asistente de investigación que toma un tema, busca artículos y publicaciones, y escribe una revisión de la literatura con citas
Qué cambia
Los agentes gestionados modifican la forma en que se construye la IA. Los desarrolladores dejan de armar la infraestructura alrededor del modelo y simplemente le dan un objetivo y un entorno aislado para trabajar. Ese entorno aislado es la abstracción, el ciclo de los agentes lo impulsa, y la API de interacciones es el único punto de entrada. Una forma sensata de comenzar es a pequeña escala, con límites estrictos sobre lo que el agente puede hacer sin supervisión, y luego escalar gradualmente.
Lecturas relacionadas
- Comparación de agentes gestionados: Anthropic, LangGraph y Google ADK — ¿Quién controla el entorno aislado, las credenciales y los registros? Una comparación de los agentes gestionados de Anthropic, LangGraph y Google ADK con información sobre precios, modos de fallo y una guía rápida para tomar decisiones.
- Una aplicación RAG mínima en Google Colab con LlamaIndex, Groq y Hugging Face — Indexa una página web con LlamaIndex, incórdiala localmente con un modelo de Hugging Face y responde preguntas mediante un LLM de código abierto alojado en Groq, todo dentro de Colab.