tenki-sandbox 0.1.0

Python SDK for Tenki Sandbox

Tenki Sandbox Python SDK

Python SDK for Tenki Sandbox: cloud sandboxes for AI agents and code execution.

pip install tenki-sandbox

from tenki_sandbox import Sandbox

with Sandbox.create(project_id="proj_123", cpu_cores=2, memory_mb=4096) as sb:
    result = sb.exec("python3", "-c", "print('hello')")
    result.check()
    print(result.stdout_text)

    sb.fs.write_text("/workspace/input.txt", "data")
    print(sb.fs.read_text("/workspace/input.txt"))

    preview = sb.expose_port(3000, ttl=3600)
    print(preview.url)

Auth

Auth resolution:

1. auth_token= passed to Client or Sandbox.create
2. TENKI_AUTH_TOKEN
3. TENKI_API_KEY

TENKI_API_ENDPOINT overrides the API URL; legacy TENKI_API_URL is also accepted.

Process API

exec collects stdout/stderr and returns a result:

result = sb.exec("npm", "test", timeout=60, env={"CI": "1"})
print(result.stdout_text)
result.check()

start returns a live process:

proc = sb.start("bash", "-lc", "read name; echo hello $name")
proc.write_stdin("tenki\n")
proc.close_stdin()
for chunk in proc.stdout:
    print(chunk.decode(), end="")
proc.wait().check()

Use shell() when you want shell parsing:

sb.shell("python3 -m http.server 3000 >/tmp/server.log 2>&1 &")

Resource APIs

from tenki_sandbox import Client, GiB

client = Client()

volume = client.volumes.create(
    workspace_id="ws_123",
    name="cache",
    size_bytes=10 * GiB,
    project_id="proj_123",
)

preview = client.preview_urls.create(
    project_id="proj_123",
    slug="demo",
    session_id=sb.id,
    port=3000,
)