Skip to content
SynthWorldDeterministic identity benchmarks
Esc
↑↓navigate↵open⌘Jpreview
On this page

Registry catalogue

Public capability and benchmark status projected from SynthWorld's resolved governance registries.

This page is generated during documentation preparation from the resolved capability and benchmark registries. It is an explicit public projection: artifact paths, digests, source locations, answer-key labels, and internal route metadata are not copied into the documentation site.

A registry entry is not publication authorization. For benchmarks, use the catalogue status and publication-gate decision shown on each record.

Benchmarks (17)

Identity-resolution ambiguity v1

  • ID: ambiguity-v1
  • Version: 1.0.0
  • Lifecycle: published
  • Kind: conformance_fixture
  • Evaluation mode: public_conformance
  • Introduced in: 0.11.0
  • Catalogue status: Approved for the docs catalogue
  • Publication decision: approved
  • Gate checks: pass: 15

Identity-resolution ambiguity v2

  • ID: ambiguity-v2
  • Version: 2.0.0
  • Lifecycle: candidate
  • Kind: generated_benchmark
  • Evaluation mode: generated_evaluation
  • Introduced in: 0.13.0
  • Catalogue status: Registry-only; no publication gate assigned
  • Publication decision: not assigned
  • Gate checks: not evaluated

Asteria C08 evidence-binding v2

  • ID: asteria-agentic-c08-v2
  • Version: 2.0.0
  • Lifecycle: candidate
  • Kind: conformance_fixture
  • Evaluation mode: public_conformance
  • Introduced in: 0.14.0
  • Catalogue status: Registry-only; no publication gate assigned
  • Publication decision: not assigned
  • Gate checks: not evaluated

Asteria Agentic v1

  • ID: asteria-agentic-v1
  • Version: 1.0.0
  • Lifecycle: published
  • Kind: conformance_fixture
  • Evaluation mode: public_conformance
  • Introduced in: 0.9.0
  • Catalogue status: Approved for the docs catalogue
  • Publication decision: approved
  • Gate checks: not_applicable: 1, pass: 14

Authority governance v1

  • ID: authority-governance-v1
  • Version: 1.0.0
  • Lifecycle: published
  • Kind: conformance_fixture
  • Evaluation mode: public_conformance
  • Introduced in: 0.13.0
  • Catalogue status: Approved for the docs catalogue
  • Publication decision: approved
  • Gate checks: pass: 15

Connection and relationship inference v1

  • ID: connection-v1
  • Version: 1.0.0
  • Lifecycle: published
  • Kind: conformance_fixture
  • Evaluation mode: public_conformance
  • Introduced in: 0.6.0
  • Catalogue status: Approved for the docs catalogue
  • Publication decision: approved
  • Gate checks: not_applicable: 2, pass: 13

Contextual access benchmark

  • ID: contextual-access
  • Version: 1.0.0
  • Lifecycle: candidate
  • Kind: generated_benchmark
  • Evaluation mode: generated_evaluation
  • Introduced in: 0.13.0
  • Catalogue status: Registry-only; no publication gate assigned
  • Publication decision: not assigned
  • Gate checks: not evaluated

Continuous assurance benchmark

  • ID: continuous-assurance
  • Version: 1.0.0
  • Lifecycle: candidate
  • Kind: generated_benchmark
  • Evaluation mode: generated_evaluation
  • Introduced in: 0.13.0
  • Catalogue status: Registry-only; no publication gate assigned
  • Publication decision: not assigned
  • Gate checks: not evaluated

Core identity world v1

  • ID: core-world-v1
  • Version: 1.0.0
  • Lifecycle: published
  • Kind: frozen_fixture
  • Evaluation mode: public_reference
  • Introduced in: 0.6.0
  • Catalogue status: Approved for the docs catalogue
  • Publication decision: approved
  • Gate checks: not_applicable: 2, pass: 13

Enterprise agentic benchmark

  • ID: enterprise-agentic
  • Version: 1.0.0
  • Lifecycle: candidate
  • Kind: generated_benchmark
  • Evaluation mode: generated_evaluation
  • Introduced in: 0.13.0
  • Catalogue status: Registry-only; no publication gate assigned
  • Publication decision: not assigned
  • Gate checks: not evaluated

Enterprise C08 evidence-binding v2

  • ID: enterprise-agentic-c08-v2
  • Version: 2.0.0
  • Lifecycle: candidate
  • Kind: conformance_fixture
  • Evaluation mode: public_conformance
  • Introduced in: 0.14.0
  • Catalogue status: Registry-only; no publication gate assigned
  • Publication decision: not assigned
  • Gate checks: not evaluated

Enterprise identity fabric profile

  • ID: enterprise-identity-fabric
  • Version: 1.0.0
  • Lifecycle: candidate
  • Kind: generated_profile
  • Evaluation mode: profile_smoke
  • Introduced in: 0.13.0
  • Catalogue status: Registry-only; no publication gate assigned
  • Publication decision: not assigned
  • Gate checks: not evaluated

PII extraction v1

  • ID: extraction-v1
  • Version: 1.0.0
  • Lifecycle: published
  • Kind: conformance_fixture
  • Evaluation mode: public_conformance
  • Introduced in: 0.6.0
  • Catalogue status: Approved for the docs catalogue
  • Publication decision: approved
  • Gate checks: not_applicable: 2, pass: 13

Households and workplaces smoke profile v1

  • ID: households-smoke-v1
  • Version: 1.0.0
  • Lifecycle: candidate
  • Kind: generated_profile
  • Evaluation mode: profile_smoke
  • Introduced in: 0.11.0
  • Catalogue status: Registry-only; no publication gate assigned
  • Publication decision: not assigned
  • Gate checks: not evaluated

Risk calibration v1

  • ID: risk-v1
  • Version: 1.0.0
  • Lifecycle: published
  • Kind: conformance_fixture
  • Evaluation mode: public_conformance
  • Introduced in: 0.6.0
  • Catalogue status: Approved for the docs catalogue
  • Publication decision: approved
  • Gate checks: not_applicable: 2, pass: 13

Search-provider projection

  • ID: search-projection
  • Version: 1.0.0
  • Lifecycle: candidate
  • Kind: projection
  • Evaluation mode: public_reference
  • Introduced in: 0.13.0
  • Catalogue status: Registry-only; no publication gate assigned
  • Publication decision: not assigned
  • Gate checks: not evaluated

Temporal broker removal benchmark

  • ID: temporal-broker-removal
  • Version: 1.0.0
  • Lifecycle: candidate
  • Kind: generated_benchmark
  • Evaluation mode: generated_evaluation
  • Introduced in: 0.13.0
  • Catalogue status: Registry-only; no publication gate assigned
  • Publication decision: not assigned
  • Gate checks: not evaluated

Capabilities (21)

Asteria agentic authority benchmark

Generate, validate, and evaluate deterministic agentic authority traces.

  • ID: agentic-authority
  • Maturity: stable
  • Since: 0.9.0
  • Support: Supported as the frozen Asteria Agentic v1 benchmark and public Python API.
  • Python interface: full (158 surfaces)
  • CLI interface: full (5 surfaces)

Asteria C08 evidence-binding v2

Generate, freeze, load, and evaluate the deterministic Asteria C08 v2 public/evaluator benchmark split.

  • ID: asteria-agentic-c08-v2
  • Maturity: preview
  • Since: 0.14.0
  • Support: Supported for repository-local candidate generation and offline evaluation while publication gates remain pending.
  • Python interface: full (69 surfaces)
  • CLI interface: none (0 surfaces)

Broker removal evaluation

Score broker removal and recurrence predictions against deterministic timelines.

  • ID: broker-evaluation
  • Maturity: preview
  • Since: 0.6.0
  • Support: Supported for the documented broker evaluation command; the top-level Python surface is currently model-only.
  • Python interface: partial (1 surfaces)
  • CLI interface: partial (1 surfaces)

Connection and relationship benchmarks

Generate and evaluate adversarial entity-resolution and relationship-inference benchmarks.

  • ID: connection-benchmark
  • Maturity: stable
  • Since: 0.6.0
  • Support: Supported deterministic connection benchmark APIs and CLI workflows.
  • Python interface: full (29 surfaces)
  • CLI interface: full (5 surfaces)

Contextual access assurance

Generate contextual access runs and validate their public execution artifacts.

  • ID: contextual-access
  • Maturity: preview
  • Since: 0.13.0
  • Support: Supported for the versioned contextual-access contract workflow.
  • Python interface: full (51 surfaces)
  • CLI interface: full (5 surfaces)

Continuous assurance

Generate and evaluate deterministic continuous-assurance scenarios.

  • ID: continuous-assurance
  • Maturity: preview
  • Since: 0.13.0
  • Support: Supported for the versioned continuous-assurance contract workflow.
  • Python interface: full (58 surfaces)
  • CLI interface: full (2 surfaces)

Core identity world

Generate deterministic synthetic identity worlds and their public identity records.

  • ID: core-identity-world
  • Maturity: stable
  • Since: 0.6.0
  • Support: Supported core-world models, generation, evaluation, and serialization APIs.
  • Python interface: full (25 surfaces)
  • CLI interface: full (2 surfaces)

Adversarial enterprise authorization

Generate and score hidden single-factor enterprise authorization counterfactuals.

  • ID: enterprise-adversarial-authorization
  • Maturity: preview
  • Since: 0.16.0
  • Support: Supported through the importable adversarial authorization Python API.
  • Python interface: full (55 surfaces)
  • CLI interface: none (0 surfaces)

Enterprise agentic assurance

Generate, validate, and evaluate fixed, smoke, standard, and longitudinal enterprise-agentic benchmark traces.

  • ID: enterprise-agentic
  • Maturity: preview
  • Since: 0.13.0
  • Support: Supported for the enterprise identity and access contract workflow.
  • Python interface: full (74 surfaces)
  • CLI interface: full (5 surfaces)

Enterprise C08 evidence-binding v2

Generate, freeze, load, and evaluate the deterministic enterprise C08 v2 public/evaluator benchmark split.

  • ID: enterprise-agentic-c08-v2
  • Maturity: preview
  • Since: 0.14.0
  • Support: Supported for repository-local candidate generation and offline evaluation while publication gates remain pending.
  • Python interface: full (64 surfaces)
  • CLI interface: none (0 surfaces)

Enterprise identity and access authoring

Scaffold, validate, and compile deterministic enterprise identity and access inputs.

  • ID: enterprise-authoring
  • Maturity: preview
  • Since: 0.13.0
  • Support: Supported for the enterprise identity and access contract workflow.
  • Python interface: full (156 surfaces)
  • CLI interface: full (3 surfaces)

Enterprise authorization scoring

Submit and independently score composed enterprise authorization decisions.

  • ID: enterprise-authorization-scoring
  • Maturity: preview
  • Since: 0.16.0
  • Support: Supported through the importable enterprise authorization Python API.
  • Python interface: full (21 surfaces)
  • CLI interface: none (0 surfaces)

Enterprise identity fabric

Generate and evaluate the deterministic enterprise identity-fabric benchmark.

  • ID: enterprise-identity-fabric
  • Maturity: preview
  • Since: 0.13.0
  • Support: Supported through the importable enterprise identity-fabric package API.
  • Python interface: full (19 surfaces)
  • CLI interface: none (0 surfaces)

Explorer v0.1

Render published Asteria Agentic v1 and verified generated enterprise-agentic smoke packages as deterministic, self-contained authority graphs with timeline replay.

  • ID: explorer-v01
  • Maturity: preview
  • Since: 0.15.0
  • Support: Preview projection contracts plus a shared packaged HTML renderer and visualization CLI for Asteria v1 and generated enterprise-agentic smoke packages; public and evaluator views remain physically separate.
  • Python interface: full (52 surfaces)
  • CLI interface: full (1 surfaces)

Exposure corpus

Generate and score deterministic breach, broker, search, and social exposure corpora.

  • ID: exposure-corpus
  • Maturity: stable
  • Since: 0.6.0
  • Support: Supported exposure-corpus models, generation, evaluation, and serialization APIs.
  • Python interface: full (14 surfaces)
  • CLI interface: full (2 surfaces)

Extraction benchmark

Generate, serialize, and evaluate exact-span extraction benchmarks.

  • ID: extraction-benchmark
  • Maturity: stable
  • Since: 0.6.0
  • Support: Supported deterministic extraction models, public inputs, answer keys, and evaluation APIs.
  • Python interface: full (25 surfaces)
  • CLI interface: full (4 surfaces)

Households profile

Generate the deterministic households profile from the command-line interface.

  • ID: households-profile
  • Maturity: preview
  • Since: 0.11.0
  • Support: Available as a CLI profile while its top-level Python API is not curated.
  • Python interface: none (0 surfaces)
  • CLI interface: full (1 surfaces)

Identity-resolution ambiguity assurance

Build and validate deterministic ambiguity evidence and run receipts.

  • ID: identity-resolution-ambiguity
  • Maturity: preview
  • Since: 0.11.0
  • Support: Supported through the importable ambiguity assurance package API.
  • Python interface: full (15 surfaces)
  • CLI interface: none (0 surfaces)

Reference baselines

Run deterministic baseline policies for selected SynthWorld benchmarks.

  • ID: reference-baselines
  • Maturity: experimental
  • Since: 0.8.0
  • Support: Reference implementations are maintained as benchmark comparators, not as production decision systems.
  • Python interface: full (7 surfaces)
  • CLI interface: none (0 surfaces)

Risk calibration benchmark

Generate and evaluate deterministic risk-calibration benchmarks.

  • ID: risk-benchmark
  • Maturity: stable
  • Since: 0.6.0
  • Support: Supported deterministic risk models, public observations, answer keys, and evaluation APIs.
  • Python interface: full (23 surfaces)
  • CLI interface: full (4 surfaces)

Shared evaluation framework

Provide common evaluation reports, metrics, and command dispatch for scoreable tasks.

  • ID: shared-evaluation-framework
  • Maturity: stable
  • Since: 0.8.0
  • Support: Supported shared evaluation report, benchmark reproduction, and command-dispatch interfaces.
  • Python interface: full (6 surfaces)
  • CLI interface: full (3 surfaces)

Was this page helpful?