DeadBolt Python client

n11-deadbolt-client connects a trusted Python executor to a separately installed DeadBolt local sidecar. It wraps a protected callback in a fresh allow/deny decision: a denial leaves the callback uncalled. The base client has no runtime dependencies. The optional openai-agents extra adapts function tools to the tested OpenAI Agents SDK contract.

This release source pairs client 2.0.0 with engine 2.0.0. Use a matching engine/client pair and check the release page and installation guide for published artifacts and package-manager availability. The wheel does not include an engine, start a sidecar or protect unwrapped routes. Earlier published packages retain their original immutable files.

Install from PyPI

Version 2.0.0 is published on PyPI. Select it explicitly in your application's virtual environment. The wheel/source archive in the GitHub release is an alternative with separately verifiable checksums:

python -m pip install 'n11-deadbolt-client==2.0.0'
# Optional SDK adapter, Python 3.10+:
python -m pip install 'n11-deadbolt-client[openai-agents]==2.0.0'

Install the matching native engine separately. No registry account or API key is required for installation. The package's description is a frozen release snapshot; current installation status is documented here.

Install a checksum-verified release wheel

Download the native ZIP, wheel and SHA256SUMS from that release. Verify the hashes and the native archive's BUILD.json revision. In your application's virtual environment:

python -m pip install /absolute/path/to/n11_deadbolt_client-2.0.0-py3-none-any.whl
/absolute/path/to/deadbolt --version
/absolute/path/to/deadbolt drill

Expected: deadbolt 2.0.0, then deadbolt drill ok. For the optional SDK adapter (Python 3.10+):

python -m pip install '/absolute/path/to/n11_deadbolt_client-2.0.0-py3-none-any.whl[openai-agents]'

The extra pins openai-agents==0.23.1. Installation does not request API credentials or make model requests. Your application configures its provider.

Build the matching release source

Use the v2.0.0 release tag once published, or a reviewed 2.0.0 source checkout, to build both components. Record git rev-parse HEAD and compare it with the release notes and the native archive's BUILD.json:

cargo build --release --locked --bin deadbolt
python -m build --outdir release-artifacts/python

Install its n11_deadbolt_client-2.0.0-py3-none-any.whl in your environment and use its target/release/deadbolt (deadbolt.exe on Windows). Keep the revision and artifact hashes with your results. A local build is not a signed download.

Version 2 fresh roots default to supervisor renewal. Provision a separate supervisor heartbeat before initial idle expiry; activity renewal is explicit operator opt-in. See supervision. Existing leases preserve their stored mode; installing a wheel does not migrate them.

Connect a protected callback

After trusted operator setup has started the sidecar and issued a scoped admission credential, configure DEADBOLT_SOCK and DEADBOLT_ADMISSION_TOKEN in the executor. Keep the operator token out of model-facing processes. Configure each lease's tools, destinations and approvals through the operator before handing it to the dispatcher.

import deadbolt_client as deadbolt

result = deadbolt.dispatch("run-001", "write_note", lambda: write_reviewed_note())
if not result["executed"]:
    show_denial(result["decision"])

write_reviewed_note and show_denial are your application's functions. Do not cache an allow or run the body after any denial or transport failure. Use the exact-action wrapper for content-bound one-shot approvals. The optional supervisor credential belongs only to trusted liveness management and cannot authorize a callback.

Guarantees and integration limits

The 2.0 engine adds explicit offline destination exceptions, fixed run deadlines and optional supervisor-only idle renewal. These require the matching engine; installing a newer client alone cannot add them to an older executable. Destination declarations are admission policy, not OS network interception. Every protected execution path must pass through fresh admission.

deadbolt_supervisor.run_owned is an optional POSIX reference for cancelling processes the trusted executor starts. It supports Linux and macOS with Python 3.13 or newer and the required wait interface. It refuses unsupported hosts, including Windows, before executing work. Ordinary callback wrappers do not cancel already running bodies. The base sidecar client supports Python 3.9+.

Guides and support

• Python setup and SDK examples
• Upgrade and Rust API migration
• Exact-action approvals
• Supervisor-only liveness
• Owned-process cancellation and its limits
• Local data and privacy
• Security reporting

DeadBolt is maintained by NextEleven LLC and licensed under Apache-2.0. Support: nextelevenstudios@gmail.com. Do not include credentials or private application data in a public issue or support report.