Skip to main content

oxp-ontology

Multi-Agent System (MAS) Execution Ontology — RDF/OWL vocabulary and Python helpers for describing, validating, and querying multi-agent system telemetry as a knowledge graph.

Part of the Observe and eXplain Platform.

What's included

Five Turtle ontologies under a single mas: namespace:

Ontology Covers
mas Core schema: MAS, Agent, LLM, Tool, Processing, their execution instances, and the normalized State/Transition trajectory
metrics Metric/MetricResult definitions and observations
semantic Embedding layer linking trajectory nodes to vector embeddings
analysis Group-scoped analysis reports (consistency, anomaly, normal behaviour)
insight Catalog-templated insights rendered onto a knowledge-graph node

Plus SHACL shapes (mas-shapes.ttl, mas-shapes-custom.ttl) for validating knowledge-graph instances against the ontology.

Install

pip install oxp-ontology

Usage

Import the ontology in your own RDF

@prefix mas: <https://outshift-open.github.io/oxp-ontology/mas#> .
@prefix ex: <http://example.org/myproject#> .

ex:myAgent a mas:Agent ;
    mas:executes ex:myTask .

ex:myMAS a mas:MAS ;
    mas:hasAgent ex:myAgent .

Access bundled files and load graphs

from oxp_ontology import get_ontology_path, load_graph

# Valid names: mas, semantic, metrics, analysis, insight,
# mas-shapes, mas-shapes-custom
mas_path = get_ontology_path("mas")

g = load_graph("mas")

Validate and verify

from oxp_ontology import validate_bundled_ontologies, verify_kg_object

# Structural validation of the bundled TTL files
report = validate_bundled_ontologies()

# SHACL-based verification of a KG object against mas-shapes(-custom).ttl
verify_kg_object(my_kg_dict)

Query with SPARQL

PREFIX mas: <https://outshift-open.github.io/oxp-ontology/mas#>

SELECT ?agent ?task WHERE {
    ?agent a mas:Agent ;
           mas:executes ?task .
}

Documentation

Full ontology documentation (Widoco reference docs and WebVOWL visualizations for each component), versioning strategy, and the source repository are at:

https://outshift-open.github.io/observe-and-explain-platform/

License

Apache-2.0. The bundled ontology files declare dcterms:license <https://opensource.org/licenses/Apache-2.0> in their headers.

Metadata

Release files for oxp-ontology 1.0.3

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

Source distribution (sdist)

Source distribution for oxp-ontology 1.0.3
File Size Uploaded Provenance Actions
oxp_ontology-1.0.3.tar.gz 34.8 kB With provenance Details for oxp_ontology-1.0.3.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for oxp-ontology 1.0.3
File Interpreter ABI Platform Actions
oxp_ontology-1.0.3-py3-none-any.whl Python 3 none any Details for oxp_ontology-1.0.3-py3-none-any.whl

Total release size: 98.0 kB

Release files / oxp_ontology-1.0.3.tar.gz

Download URL oxp_ontology-1.0.3.tar.gz
Size 34.8 kB
Tags Source
SHA-256 checksum
How to use checksums
9ea6e442560c1e45ff4ab8e82284701021c4ff5a1045dca8b238924228f87de7
BLAKE2b-256 checksum
How to use checksums
bd9cfd840ea457be7df0bf6ac39024dea523f5eb7a7d6a4e9d66fe7d7fa1f857
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

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.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 12, 2026.

Transparency log

Release files / oxp_ontology-1.0.3-py3-none-any.whl

Download URL oxp_ontology-1.0.3-py3-none-any.whl
Size 63.2 kB
Tags Python 3
SHA-256 checksum
How to use checksums
6295769792662052e95b5d253a664917c2c6cd9e2128ae5d1eefdd1ace984480
BLAKE2b-256 checksum
How to use checksums
b21b5217733d40a0b036a10a6e67050d9b5c60681c600ad174b07086c1f11f1a
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

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.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 12, 2026.

Transparency log

Security for oxp-ontology 1.0.3

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.

Full provenance

Both files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for oxp_ontology-1.0.3.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 12, 2026.

Transparency log

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.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

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