Metadata-Version: 2.4
Name: sectum-ai
Version: 0.1.6
Summary: Sectum AI - multi-tenant AI verification: core substrate runner and the sectum-ai CLI.
Project-URL: Homepage, https://sectum.ai
Project-URL: Documentation, https://docs.sectum.ai
Project-URL: Repository, https://github.com/sectum-ai/sectum-ai
Project-URL: Changelog, https://github.com/sectum-ai/sectum-ai/blob/main/CHANGELOG.md
Author: Sectum AI
License-Expression: Apache-2.0
Keywords: ai-security,llm,multi-tenant,rag,verification
Classifier: Development Status :: 2 - Pre-Alpha
Classifier: Intended Audience :: Developers
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3.12
Classifier: Topic :: Security
Requires-Python: >=3.12
Requires-Dist: pydantic>=2.9
Requires-Dist: pyyaml>=6
Requires-Dist: sectum-ai-adapters
Requires-Dist: sectum-ai-evidence
Requires-Dist: sectum-ai-probes
Requires-Dist: sectum-ai-spec
Requires-Dist: typer>=0.12
Provides-Extra: encryption
Requires-Dist: cryptography>=43; extra == 'encryption'
Provides-Extra: openai
Requires-Dist: openai>=1.40; extra == 'openai'
Provides-Extra: sentence-transformers
Requires-Dist: sentence-transformers>=3; extra == 'sentence-transformers'
Provides-Extra: weasyprint
Requires-Dist: sectum-ai-evidence[weasyprint]; extra == 'weasyprint'
Description-Content-Type: text/markdown

# sectum-ai

**Multi-tenant AI verification.** This is the core distribution of [Sectum AI](https://github.com/sectum-ai/sectum-ai):
the marker-substrate runner and the `sectum-ai` command-line interface.

Sectum AI provisions synthetic tenants on an AI stack, seeds them with
cryptographic canary markers, runs benign and adversarial probes from each
tenant's session, and detects cross-tenant data leakage across every surface —
producing tamper-evident, control-mapped evidence that an auditor accepts.

## Install

```sh
pip install sectum-ai
```

This pulls the full family: `sectum-ai-spec` (data models), `sectum-ai-probes`
(the Class 1–11 attack catalog + leak-detection pipeline), `sectum-ai-adapters`
(connectors for vector stores, caches, observability, RAG, agents, and MCP), and
`sectum-ai-evidence` (the tamper-evident evidence chain + `sectum-ai verify`).

## Quickstart

```sh
sectum-ai seed      # provision synthetic tenants + plant canary markers
sectum-ai probe     # run the attack catalog from each tenant's session
sectum-ai report    # assemble a signed, control-mapped evidence pack (JSON + PDF)
sectum-ai verify .sectum-ai/evidence.json   # independently re-verify the pack
```

## Links

- Documentation: <https://docs.sectum.ai>
- Source, full README, and attack catalog: <https://github.com/sectum-ai/sectum-ai>

Apache-2.0. The marker substrate, attack catalog, adapters, evidence chain, and
the independent `sectum-ai verify` are fully open source.
