Skip to main content

DOI

xkernel

Invariant Deterministic Kernel (k = 1, Boolean Closure)

xkernel is a minimal, invariant execution kernel.

It defines:

  • a single admissible action quantum (k = 1),
  • deterministic replay semantics,
  • boolean closure against a target state,
  • canonical execution hashing,
  • and tamper-evident receipts.

xkernel is NOT a runtime engine, simulator, or pattern system. It has NO kernel object, NO mutable configuration, and NO hidden state.

The kernel is the law. Executions and receipts are the artifacts.


What xkernel guarantees

  • Determinism Replaying the same execution always yields the same final state.

  • Admissibility Any step with action != 1 is invalid.

  • Closure Closure is a boolean: an execution either closes to a target state or it does not.

  • Canonical identity Executions and receipts have stable, content-addressed identifiers:

    • xk:sha256:... for executions
    • xr:sha256:... for receipts
  • Tamper evidence Any mutation of execution or receipt content changes its hash and fails verification.


What xkernel does NOT do

  • No engine orchestration
  • No filaments, fibers, or bundles
  • No pattern detection or lenses
  • No identity, signing, or trust semantics
  • No backward compatibility with v1

Those concerns belong in downstream systems.


Installation

pip install -e .

Command-line interface

xkernel ships with a strict reference CLI.

xkernel --help

Available commands:

  • validate Validate execution admissibility and replay
  • hash Compute execution content address (xk:sha256:...)
  • receipt Emit receipt JSON for an execution (optional target closure)
  • verify Verify receipt integrity against an execution
  • receipt-hash Compute receipt content address (xr:sha256:...)

Example:

xkernel validate exec.json
xkernel hash exec.json
xkernel receipt exec.json --target target.json > receipt.json
xkernel verify receipt.json exec.json

Core model (conceptual)

StateVector  : immutable integer coordinate vector
Step         : delta + action (must equal k = 1)
Execution    : init + ordered steps + final
Closure      : boolean equality to a target under replay
Receipt      : canonical record of execution + closure

There is no kernel instance. Correctness is enforced structurally.


Testing

pytest

All tests assert invariants, determinism, hashing stability, and tamper detection.


Versioning

v2.0.0 introduces invariant semantics and removes all v1 engine concepts. There is no backward compatibility.


License

MIT - see LICENSE.

Metadata

Release files for xkernel 2.0.1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for xkernel 2.0.1
File Size Uploaded Provenance
xkernel-2.0.1.tar.gz 11.7 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for xkernel 2.0.1
File Interpreter ABI Platform
xkernel-2.0.1-py3-none-any.whl Python 3 none any Details

Total release size: 23.0 kB

Release files / xkernel-2.0.1.tar.gz

Download URL xkernel-2.0.1.tar.gz
Size 11.7 kB
Tags Source
SHA-256 checksum
How to use checksums
9c325400a97670297b4437170256dd47510541b59ec85ad1a73cb67c3d9bbd6e
BLAKE2b-256 checksum
How to use checksums
185c6de199f5f6b2c0ef55b279e0a024357f05f648f89b595d75c8869c66da78
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.13.7

Release files / xkernel-2.0.1-py3-none-any.whl

Download URL xkernel-2.0.1-py3-none-any.whl
Size 11.3 kB
Tags Python 3
SHA-256 checksum
How to use checksums
b35b148b1d1e05a02a5bbf88a21f9130db53ebbf460549d9b0d458238ccd2c19
BLAKE2b-256 checksum
How to use checksums
bfaf40e0894238729e4d527be299f4dfac38de65128bfb1571302086e7d6aae9
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.2.0 CPython/3.13.7

Security for xkernel 2.0.1

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page