Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions .env.example
Original file line number Diff line number Diff line change
Expand Up @@ -6,4 +6,6 @@ OPENAI_ORG = "org-123"
BROWSERBASE_API_KEY="00000000-0000-0000-0000-000000000000"
BROWSERBASE_PROJECT_ID="bb_live_00000000-00000"

E2B_API_KEY="e2b_key"

SCRAPYBARA_API_KEY="scrapy-123"
2 changes: 2 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -28,6 +28,7 @@ Other included sample [computer environments](#computer-environments):

- [Docker](https://docker.com/) (containerized desktop)
- [Browserbase](https://www.browserbase.com/) (remote browser, requires account)
- [E2B](https://e2b.dev) (remote computer, open-source, requires account)
- [Scrapybara](https://scrapybara.com) (remote browser or computer, requires account)
- ...or implement your own `Computer`!

Expand Down Expand Up @@ -94,6 +95,7 @@ This sample app provides a set of implemented `Computer` examples, but feel free
| `LocalPlaywright` | local-playwright | `browser` | Local browser window | [Playwright SDK](https://playwright.dev/) |
| `Docker` | docker | `linux` | Docker container environment | [Docker](https://docs.docker.com/engine/install/) running |
| `Browserbase` | browserbase | `browser` | Remote browser environment | [Browserbase](https://www.browserbase.com/) API key in `.env` |
| `E2B` | e2b | `linux` | Open-source desktop environment | [E2B](https://e2b.dev) API key in `.env` |
| `ScrapybaraBrowser` | scrapybara-browser | `browser` | Remote browser environment | [Scrapybara](https://scrapybara.com/dashboard) API key in `.env` |
| `ScrapybaraUbuntu` | scrapybara-ubuntu | `linux` | Remote Ubuntu desktop environment | [Scrapybara](https://scrapybara.com/dashboard) API key in `.env` |

Expand Down
1 change: 1 addition & 0 deletions computers/config.py
Original file line number Diff line number Diff line change
Expand Up @@ -7,4 +7,5 @@
"browserbase": BrowserbaseBrowser,
"scrapybara-browser": ScrapybaraBrowser,
"scrapybara-ubuntu": ScrapybaraUbuntu,
"e2b": E2BDesktop,
}
1 change: 1 addition & 0 deletions computers/contrib/__init__.py
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
from .e2b import E2BDesktop
86 changes: 86 additions & 0 deletions computers/contrib/e2b.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,86 @@
import base64
from e2b_desktop import Sandbox
from typing import Literal

class E2BDesktop:
"""
E2B Desktop is an open-source desktop environment for AI Agents.
You can get started for free at https://e2b.dev or read our docs at https://docs.e2b.dev
"""

def __init__(self):
self.environment: Literal["windows", "mac", "linux", "browser"] = "linux" # "windows", "mac", "linux", or "browser"
self.dimensions = (1024, 768)
self.stream_url: str | None = None

def __enter__(self):
print("Starting E2B Desktop Sandbox")
self.sandbox = Sandbox(
resolution=self.dimensions,
timeout=300,
)

print(f"Started E2B Desktop Sandbox with id '{self.sandbox.sandbox_id}'")

self.sandbox.stream.start(require_auth=True)

return self

def __exit__(self, exc_type, exc_val, exc_tb):
self.sandbox.kill()

def screenshot(self) -> str:
screenshot = self.sandbox.screenshot(format="bytes")
base64_image = base64.b64encode(screenshot).decode("utf-8")
return base64_image

def click(self, x: int, y: int, button: str = "left") -> None:
match button:
case "left":
self.sandbox.left_click(x, y)
case "right":
self.sandbox.right_click(x, y)
case "middle":
self.sandbox.middle_click(x, y)

def double_click(self, x: int, y: int) -> None:
self.sandbox.double_click(x, y)

def scroll(self, x: int, y: int, scroll_x: int, scroll_y: int) -> None:
self.sandbox.move_mouse(x, y)
if scroll_y < 0:
self.sandbox.scroll("up", abs(scroll_y))
elif scroll_y > 0:
self.sandbox.scroll("down", scroll_y)

def type(self, text: str) -> None:
self.sandbox.write(text)

def wait(self, ms: int = 1000) -> None:
self.sandbox.wait(ms)

def move(self, x: int, y: int) -> None:
self.sandbox.move_mouse(x, y)

def keypress(self, keys: list[str]) -> None:
self.sandbox.press(keys)

def drag(self, path: list[dict[str, int]]) -> None:
if not path:
return
start_x = path[0]["x"]
start_y = path[0]["y"]

end_x = path[-1]["x"]
end_y = path[-1]["y"]

self.sandbox.drag((start_x, start_y), (end_x, end_y))

def get_environment(self) -> Literal["windows", "mac", "linux", "browser"]:
return self.environment

def get_dimensions(self) -> tuple[int, int]:
return self.dimensions

def get_current_url(self) -> str:
return self.sandbox.stream.get_url(auth_key=self.sandbox.stream.get_auth_key())
92 changes: 92 additions & 0 deletions computers/e2b.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,92 @@
import base64
from e2b_desktop import Sandbox

cua_e2b_key_mapping = {
"ENTER": "Return",
"LEFT": "Left",
"RIGHT": "Right",
"UP": "Up",
"DOWN": "Down",
"ESC": "Escape",
"SPACE": "space",
"BACKSPACE": "BackSpace",
"TAB": "Tab",
}

class E2BDesktop:
"""
E2B Desktop is a desktop environment in the cloud.
"""

def __init__(self):
self.environment = "linux" # "windows", "mac", "linux", or "browser"
self.dimensions = (1024, 768)

def __enter__(self):
print("Starting E2B Desktop Sandbox")
self.sandbox = Sandbox(
resolution=self.dimensions,
timeout=300,
)

print(f"Started E2B Desktop Sandbox with id '{self.sandbox.sandbox_id}'")

print("Starting Desktop Stream")
self.sandbox.stream.start(require_auth=True)

stream_url = self.sandbox.stream.get_url(auth_key=self.sandbox.stream.get_auth_key())
print(f"Desktop Stream is running at {stream_url}")
return self

def __exit__(self, exc_type, exc_val, exc_tb):
self.sandbox.kill()

def screenshot(self) -> str:
screenshot = self.sandbox.screenshot()
base64_image = base64.b64encode(screenshot).decode("utf-8")
return base64_image

def click(self, x: int, y: int, button: str = "left") -> None:
match button:
case "left":
self.sandbox.move_mouse(x, y)
self.sandbox.left_click()
case "right":
self.sandbox.move_mouse(x, y)
self.sandbox.right_click()
case "middle":
self.sandbox.move_mouse(x, y)
self.sandbox.middle_click()

def double_click(self, x: int, y: int) -> None:
self.sandbox.move_mouse(x, y)
self.sandbox.double_click()

def scroll(self, x: int, y: int, scroll_x: int, scroll_y: int) -> None:
self.sandbox.move_mouse(x, y)
self.sandbox.scroll(scroll_x, scroll_y)

def type(self, text: str) -> None:
self.sandbox.write(text)

def wait(self, ms: int = 1000) -> None:
self.sandbox.wait(ms)

def move(self, x: int, y: int) -> None:
self.sandbox.move_mouse(x, y)

def keypress(self, keys: list[str]) -> None:
# convert keys to e2b desktop keys
mapped_keys = [cua_e2b_key_mapping.get(key, key) for key in keys]
self.sandbox.press(mapped_keys)

def drag(self, path: list[dict[str, int]]) -> None:
if not path:
return
start_x = path[0]["x"]
start_y = path[0]["y"]

end_x = path[-1]["x"]
end_y = path[-1]["y"]

self.sandbox.drag((start_x, start_y), (end_x, end_y))
1 change: 1 addition & 0 deletions requirements.txt
Original file line number Diff line number Diff line change
Expand Up @@ -21,3 +21,4 @@ scrapybara>=2.3.6
sniffio==1.3.1
typing_extensions==4.12.2
urllib3==2.3.0
e2b_desktop==1.7.1