Integrations
Alice decides; enforcement points enforce. This page says exactly what exists today. Vendor names appear only as examples of an enforcement-point pattern; no partnership, endorsement, or integration is implied by a name.
Running in production today.
Implemented and tested in this codebase; not yet in the live request path.
Runs on synthetic data on this site.
Interface defined; needs a counterpart to connect.
On the roadmap; not built.
An idea or pattern; nothing exists.
| Component | Role | Status | What it does | Basis for the label |
|---|---|---|---|---|
| Nexus Intelligence feeds | SIGNAL | LIVE | Regulatory, breach, and workforce signals that raise policy proposals. | Public terminal is live; the policy-proposal linkage is demonstrated on synthetic policies. |
| Nexus Readiness assessment | SIGNAL | LIVE | Critical gaps that cap autonomy or add review requirements. | Scoring engine is live for signed-in organizations; the public linkage uses the synthetic assessment. |
| Model eligibility evaluator | PDP | SUPPORTED | Deterministic model verdicts with reason codes from the governed registry. | Implemented and tested in shadow mode; not yet in the live inference path. |
| Trust Receipt hash chain | EVIDENCE | SUPPORTED | Append-only, tamper-evident chain and evidence package. | Implemented and tested; production persistence is an approval-gated step. |
| Alice control plane (policy engine + gateway) | PDP | DEMO | Default-deny policy engine (BLOCK > HOLD > ALLOW), agent identity with hashed secrets, config-pinned audit ledger. | Exists as a tested Python kernel (v0.2) in this organization's repository; not wired into the Nexus request path. The web decision engine mirrors its rules. |
| Nexus decision engine (this site) | PDP | DEMO | Structured ALLOW / REVIEW / BLOCK evaluation with checks, reusing the model-eligibility evaluator. | Runs over synthetic Lakeview requests; deterministic and tested. Not enforcing live traffic. |
| Hospital SSO / IAM | IDENTITY | PLANNED | Human identity verification for every governed request. | Not integrated; the demo uses a synthetic verified identity. |
| MCP gateway | PEP | PLANNED | Enforce tool permissions and envelopes for agent tool calls at the gateway. | Design target for the Alice gateway; no gateway is deployed. |
| API gateway / application proxy | PEP | PLANNED | Route model requests through Alice before any provider is called. | The residency-governance module exists but is not wired to the provider path. |
| DLP / DSPM (for example Forcepoint Data Security Cloud) | PEP | CONCEPTUAL | Data-loss prevention and data-security posture tools could act as enforcement points for Alice decisions on data movement. | No Forcepoint or other DLP integration exists. Named as an example of the enforcement-point pattern only; no partnership or endorsement is implied. |
| Nexus gateway | PEP | CONCEPTUAL | A Nexus-hosted enforcement point combining identity, model routing, and receipts. | Roadmap concept. |
Alice on this site runs over the synthetic Lakeview General Hospital environment. Verdicts, checks, and receipts are computed by the Nexus decision engine on synthetic requests; nothing here is a customer, a patient, or a production deployment. Signed-in organizations see their own agents, decisions, and receipts.