Skip to main content
ZR
Toggle navigation
Work

Flagship mandate

PolicyStrata

Release evidence for LLM data agents before they touch governed systems.

Tier
Flagship mandate
Pillar
Agent governance
Visibility
Public
Status
functional
Date
Jun 2026

Case boundary

Python CLI and research artifact for testing cross-layer policy drift across manifests, semantic plans, SQL compilers, databases, release filters, and scanner traces.

Claim boundary

Research and regression evidence, not an authorization boundary or production recall claim.

PolicyStrata is a deterministic regression-testing framework for policy drift in LLM data-agent stacks.

It generates principals, requests, semantic plans, database states, lowered queries, and release decisions, then compares each layer against a canonical policy oracle. Failures are minimized into small reproducible witnesses.

System

  • Canonical policy oracle kept independent from manifests, validators, compilers, and database behavior
  • Deterministic support, finance, and ClickHouse-style benchmark domains
  • Scanner mode for imported application traces and release-gate workflows
  • Packaged Python CLI, PyPI distribution, GitHub Action, Docker fixtures, and artifact bundle

Runtime

  • policystrata run writes traces, summaries, metadata, manifests, and minimized witnesses
  • policystrata scan checks exported SQL/tool-call traces against a scanner config
  • Frozen benchmark manifests protect held-out suites from accidental detector drift
  • Reproduction scripts rebuild paper-facing evidence tables from a clean checkout

Proof

  • Paper artifact reports 1720/1720 injected non-clean cases with 0 false positives on 80 clean controls
  • Failing runs minimize to witness JSON that localizes the violated layer and obligation
  • Tests cover policy logic, compiler behavior, database assets, scanner config, evidence output, and benchmark freezing
  • Public release includes SHA256-pinned paper and artifact archive