sectum-ai-spec
Shared data models and JSON Schema for Sectum AI, the multi-tenant AI verification toolkit.
This distribution holds the Pydantic v2 models and exported JSON Schema that
every other Sectum package builds on — Scenario, Marker,
GroundTruthManifest, ProbeStep, Observation, Finding, RunResult, and
EvidencePack — plus the typed error hierarchy (SectumError and friends).
It is the lowest layer in the package graph and depends on nothing else in the
family.
pip install sectum-ai-spec
Most users install the umbrella package sectum-ai
instead, which pulls this in automatically.
JSON Schema
Every model is also published as a standalone JSON Schema document under
sectum_ai/spec/schemas/<Model>.schema.json (shipped in the wheel). Each carries a
$schema dialect (draft 2020-12) and a version-pinned $id
(https://schemas.sectum.ai/<schema_version>/<Model>.schema.json), so external
tooling can validate Sectum artifacts without importing Python:
from sectum_ai.spec import json_schemas
from sectum_ai.spec.schema import SCHEMA_DIR # the committed, packaged artifacts
finding_schema = json_schemas()["Finding"]
The Pydantic models are authoritative; the schemas are generated from them.
Regenerate after a model change with uv run python scripts/gen_schemas.py
(a test fails if a committed artifact drifts). The schema_version field
versions the models — bump it for a backward-incompatible change, then regenerate.
- Documentation: https://docs.sectum.ai
- Source: https://github.com/sectum-ai/sectum-ai
Apache-2.0.
Metadata
Release files for sectum-ai-spec 0.10.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| sectum_ai_spec-0.10.1.tar.gz | 27.8 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| sectum_ai_spec-0.10.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 70.3 kB
Release files / sectum_ai_spec-0.10.1.tar.gz
| Download URL | sectum_ai_spec-0.10.1.tar.gz |
|---|---|
| Size | 27.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
f671d8cb5a46f0dfdcf690c394ce97a7b01a16378838830d3181aa5e38adc3b4
|
|
BLAKE2b-256 checksum How to use checksums |
7a4bd536c3749e5ebce088a4296761eda68d38dacd78345fa68af8cddf203f4b
|
| 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 Sep 2, 2026.
Transparency logRelease files / sectum_ai_spec-0.10.1-py3-none-any.whl
| Download URL | sectum_ai_spec-0.10.1-py3-none-any.whl |
|---|---|
| Size | 42.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
4a528088c8131459df55bfd696a8f6c572d0ba8a26297b4f685cb7222523f4a6
|
|
BLAKE2b-256 checksum How to use checksums |
30c2deaf9310fd1a73c4dad95ffaeba08a81c492b2366db7d7b3297df9d875ac
|
| 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 Sep 2, 2026.
Transparency log