Python SDK

Install speedle 0.1.0 with pip or uv, then call is_allowed and check against a local Speedle or Speedle Cloud.

The Python SDK needs Python 3.10 or later; its only runtime dependency is httpx. It has a sync Client and an asyncio AsyncClient with the same arguments. Package: speedle 0.1.0 on PyPI. Reference: speedle-sdk/python.

The examples check the policy described in SDKs.

Add the dependency

With pip, in a virtual environment:

python3 -m venv .venv
. .venv/bin/activate
pip install speedle==0.1.0

Or with uv, in a uv project:

uv add speedle==0.1.0

Call a local Speedle

Start Speedle and create the example policy as described in SDKs. Save this program as main.py:

import sys

from speedle import Auth, Client, SpeedleError

try:
    with Client(
        backend="speedle-plus",
        endpoint="http://localhost:6734",
        auth=Auth.none(),
        service_name="docs",
    ) as speedle:
        allowed = speedle.is_allowed("alice", "get", "/docs/public")
        print(f"alice get /docs/public: allowed={allowed}")

        decision = speedle.check("alice", "delete", "/docs/public")
        print(f"alice delete /docs/public: allowed={decision.allowed} reason={decision.reason!r}")
except SpeedleError as e:
    print(f"Speedle error: {e}", file=sys.stderr)
    sys.exit(1)

Run it with python main.py in the virtual environment, or with uv:

uv run main.py

Output:

alice get /docs/public: allowed=True
alice delete /docs/public: allowed=False reason=<Reason.NO_APPLICABLE_POLICIES: 3>

Call Speedle Cloud

Speedle Cloud is coming soon.

The program reads the service name from SPEEDLE_SERVICE and the service’s token from SPEEDLE_TOKEN. SPEEDLE_ENDPOINT overrides the default endpoint https://cloud.speedle.io. Replace main.py with:

import os
import sys

from speedle import Auth, Client, SpeedleError

try:
    with Client(
        backend="speedle-cloud",
        endpoint=os.environ.get("SPEEDLE_ENDPOINT", "https://cloud.speedle.io"),
        auth=Auth.bearer(os.environ["SPEEDLE_TOKEN"]),
        service_name=os.environ["SPEEDLE_SERVICE"],
    ) as speedle:
        allowed = speedle.is_allowed("alice", "get", "/docs/public")
        print(f"alice get /docs/public: allowed={allowed}")

        decision = speedle.check("alice", "delete", "/docs/public")
        print(f"alice delete /docs/public: allowed={decision.allowed} reason={decision.reason!r}")
except SpeedleError as e:
    print(f"Speedle error: {e}", file=sys.stderr)
    sys.exit(1)

Set the environment and run it. read -rs keeps the token out of your shell history and screen:

export SPEEDLE_SERVICE=docs
read -rs SPEEDLE_TOKEN && export SPEEDLE_TOKEN
python main.py

Output, when the service has the example policy:

alice get /docs/public: allowed=True
alice delete /docs/public: allowed=False reason=<Reason.NO_APPLICABLE_POLICIES: 3>

The token is a server-side secret. Never ship it to browsers, mobile apps or desktop clients.