SynthWorld 0.17.0

Added

• Generated enterprise-agentic standard and longitudinal tiers (#27). A new
  independently versioned V2 configuration and artifact family adds deterministic
  multi-organisation standard scale plus a 180-day longitudinal schedule covering
  repeated credential rotation, joiner/mover/leaver changes, suspension, policy
  activation, agent offboarding with active credentials, revocation propagation,
  evidence retention, and audit-time scoring. The released smoke V1 and frozen
  Asteria/C08 contracts remain unchanged. Public topology, opaque credential-handle,
  and lifecycle projections stay physically separated from evaluator-only case
  labels and truth; generic
  loaders, CLI config/public-only export, derived denominated integrity metrics,
  end-to-end observed-action scoring, and an external digest-bound runtime/memory
  receipt complete the generated-scale path. Stress remains explicitly deferred to
  generic scale issue #3 and generated tiers remain workloads rather than vendor
  leaderboard claims or mandatory package fixtures.
• Enterprise agentic identity policy pilot. A repository example now compares
  experiment-owned RBAC, ABAC, ReBAC, and default-deny combined policy views over
  one generated enterprise-agentic smoke world. The policy runner consumes only
  the verified public tree and writes digest-bound decision traces; a separate
  scorer verifies the matching evaluator tree and emits independent metrics plus
  deterministic, visibly watermarked comparison HTML. A new end-to-end guide
  covers generation, public projection and layout JSON, the evaluator overlay,
  Explorer HTML, replay, scoring, reproducibility, and the boundary between this
  teaching pilot and SynthWorld authorization contracts or production enforcement.
• Generated enterprise-agentic Explorer adapter (#149). The synthworld visualize command now renders a verified released generated
  enterprise-agentic smoke package as deterministic, self-contained HTML when
  selected explicitly with --package-profile generated-enterprise-agentic. An
  independently versioned enterprise-agentic-generated-v1 projection records
  the generated configuration digest, world identity, tier, and public
  artifact-set digest; a new 1.0.0 generated layout contract computes a
  deterministic kind-layered grid from the projection alone, so no frozen
  1.0.0 Explorer or Asteria contract widens in place and published Asteria
  HTML bytes are unchanged. The public path consumes only the verified public
  tree; evaluator truth requires the separate evaluator tree, is digest-bound
  through the reused overlay contract, and renders visibly watermarked.
  Unsupported generated tiers or package versions fail explicitly, and the
  shared #52 renderer, assets, and interaction are reused without a second
  viewer implementation.
• Packaged Asteria Explorer renderer (#52). The synthworld visualize command
  now renders the checksum-verified published Asteria Agentic v1 public package as
  deterministic, self-contained HTML with an interactive authority graph, chain
  inspection, and timeline replay. Cytoscape and build-time ELK versions are pinned,
  generated assets and layout coordinates are checksum-bound, no runtime network
  resources are loaded, and dependency notices ship in the wheel. The new layout
  2.0.0 contract explicitly records world seed, world schema version, and
  visualisation profile identity while preserving layout 1.0.0 unchanged. Evaluator
  truth requires a separately verified evaluator package and produces prominently
  watermarked output; public HTML contains no evaluator overlay. Generated
  enterprise-agentic support is supplied by the separately versioned adapter
  described above; enterprise authorization packages remain outside Explorer.

What's Changed

• Publish frozen enterprise authorization experiment evidence by @bluntmachetti in #146
• Complete the packaged Asteria Explorer renderer by @bluntmachetti in #152
• Document the OPA/AuthZEN enterprise experiment by @bluntmachetti in #148
• Add an Explorer adapter for generated enterprise-agentic worlds (#149) by @bluntmachetti in #153
• Add an enterprise agentic identity policy pilot by @bluntmachetti in #154
• Prepare SynthWorld 0.17.0 release by @bluntmachetti in #155

Full Changelog: v0.16.0...v0.17.0

SynthWorld 0.16.0

Added

• Stable enterprise authorization consumer API (#139). A curated
  synthworld.enterprise.consumer namespace now exposes the authoring rows,
  vocabularies, limits, artifact/result wrappers, compilers, prediction rows,
  evaluators, and exact canonical digest helpers needed for a released-wheel
  workflow. A versioned operator-side compiler provenance artifact maps canonical
  topology and directory-policy locations to compiled opaque IDs without entering
  public product input or evaluator truth. Authorization public bundles now carry
  their evaluation scope, and an isolated-wheel test exercises topology import,
  all three mechanism families, composition, public-only prediction, and scoring.
• Publicly solvable composed enterprise authorization scoring (#137). New
  independently versioned public evaluation-scope, system submission, and
  evaluation-report contracts score effective and final decisions separately
  from RBAC, ABAC, ReBAC, conflict, binding, lifecycle, and runtime-gate
  outcomes. Exact public cell inventories, artifact and schema bindings,
  deterministic system/policy metadata, and per-metric denominators are
  enforced without changing existing 1.0.0 schemas or golden bytes.
• Discriminating adversarial enterprise authorization cases (#138). A new
  independently versioned reference profile adds hidden single-factor pairs for
  tenant, scope, credential binding, time, clearance, and RBAC/ReBAC authority
  composition. Its public policy supports generic tenant inequality without
  enumerating negative target IDs, candidate action attempts remain separate
  from persistent grants, and opaque attempt IDs reveal no pair or verdict
  labels. Independent metrics report total cohorts and discriminating
  denominators; seven weak baselines fail their dedicated dimensions without
  changing frozen schemas or golden artifacts.
• Published Phase 1 and Phase 2 enterprise authorization evidence. The
  GitHub release carries checksum-bound historical, reproduction, and reference
  archives from the Topaz experiment conducted against SynthWorld 0.15.0. The
  signed release source records the expected digests and the documentation states
  which archives contain evaluator evidence. These retained artifacts do not
  retroactively rescore the experiment with the new 0.16.0 contracts.

What's Changed

• Clarify Explorer 0.15.0 release boundary by @bluntmachetti in #136
• Add composed enterprise authorization scoring by @bluntmachetti in #141
• Stabilize the released enterprise authorization consumer API by @bluntmachetti in #143
• Document enterprise authorization reproduction context by @bluntmachetti in #144
• Add adversarial enterprise authorization cases by @bluntmachetti in #142
• Prepare SynthWorld 0.16.0 by @bluntmachetti in #145

Full Changelog: v0.15.0...v0.16.0

Enterprise authorization Topaz reference experiment (SynthWorld 0.16.0)

Frozen enterprise authorization reference experiment

This is a frozen, unsupported record of one experiment using
idcognito-synthworld==0.16.0 and Topaz 0.33.16. It is evidence of what was run,
not a maintained adapter, product certification, vendor comparison, or general
authorization-correctness claim.

The experiment generated a deterministic enterprise identity/authorization world,
projected only public artifacts into Topaz, captured raw decisions, sealed typed
submissions before scoring, and opened evaluator truth only in a separate offline
scorer capability.

Recorded results

• 1,699 objects and 5,879 relations loaded and read back exactly;
• 3,209/3,209 Britannia decisions and 14/14 adversarial decisions normalized;
• 40/40 structural, isolation, integrity, and presentation checks passed;
• 9/9 deliberately faulty authorization controls detected;
• 4/4 unsealed, mutated, cross-artifact, and version-mismatched submissions refused;
• no aggregate score was calculated; unexercised and not-publicly-winnable metrics
  remain explicit in the report.

Assets

• enterprise-authorization-topaz-reproduction-kit-0.16.0-1.zip
  (sha256:16071b56892d39542817b72401991adde38f878184c7dbb7b274662b24aa4b5f):
  source, pinned dependencies, topology, policies, adapters, documentation, and the
  one-command workflow. It contains no pre-generated evaluator artifacts.
• enterprise-authorization-topaz-reference-run-0.16.0-1.zip
  (sha256:202d1377f1f07bed57b82fc829f56082f23ffe2af626f5a8aeb39093ed0bddbe):
  the source plus retained public artifacts, separate evaluator artifacts, raw and
  normalized Topaz responses, sealed submissions, reports, controls, and the
  public-only HTML projection.
• ASSET-METADATA.json binds the experiment, source commit, source tree, package
  version, asset sizes, asset digests, and result counts.
• SHA256SUMS verifies every custom asset.

Reproduce

sha256sum -c SHA256SUMS
unzip enterprise-authorization-topaz-reproduction-kit-0.16.0-1.zip
cd enterprise-authorization-topaz-reproduction-kit-0.16.0-1
bin/run_lab.sh

The reproduction ZIP was itself executed from a fresh extraction before this
release was published and passed all 40 checks. First execution requires network
access to obtain digest-pinned images and hash-pinned packages; reruns can use the
retained local cache.

Community experiments are owned and supported by their authors. They may be
listed as self-reported results in the SynthWorld
Experiment results
Discussion category. A listing is not review, reproduction, endorsement, or a
support commitment by SynthWorld maintainers.

Enterprise authorization OPA/AuthZEN reference experiment (SynthWorld 0.16.0, revision 2)

Frozen enterprise authorization OPA/AuthZEN reference experiment

This is the authoritative publication of agent-auth-3, a frozen, unsupported
external-consumer experiment using idcognito-synthworld==0.16.0, Open Policy
Agent 1.4.2, and an experiment-owned adapter for OpenID Authorization API 1.0.

Publication revision 2 supersedes
enterprise-authorization-opa-authzen-0.16.0-1. Revision 1 accidentally retained
unmanifested Python bytecode caches in both ZIPs. This revision removes only those
generated cache files. Experiment inputs, source files, retained evidence, results,
and internal checksums are unchanged. The immutable revision 1 remains available
as a transparent record of the correction.

This experiment is evidence of what was run. It is not a maintained adapter,
AuthZEN conformance claim, product certification, vendor comparison, production
enforcement result, or general authorization-correctness claim.

The experiment maps two fictional organization topologies using one deterministic
method, projects only physically separated public artifacts into OPA, binds every
AuthZEN request entity and evaluation-context field to its public benchmark cell,
seals submissions before scoring, and opens evaluator truth only in the offline
scorer capability.

Recorded results

• two topology lanes and 649 evaluation cells;
• 649/649 effective authorization decisions correct;
• 648/649 final decisions correct;
• 20/20 deliberately faulty policy and adapter controls detected;
• 16/16 fixed-cell-ID AuthZEN request mutations refused;
• 20/20 missing, modified, cross-topology and cross-version seal controls passed;
• 175/175 deterministic files byte-identical across two complete runs;
• a fresh extraction of this revision's reproduction ZIP completed successfully;
• every release and retained-reference checksum verified.

The residual binding errors are reported rather than hidden: canonical
account-to-principal binding is evaluator-only, so plausible same-kind substitutions
are not always decidable from public evidence.

Assets

• enterprise-authorization-opa-authzen-0.16.0-2-reproduction-kit.zip
  (sha256:8a48b643dfaeb84d3d0ec5ecc0683c02cea109dbc73e4404928f91f4dfea9cfa):
  source, pinned dependencies, two topology inputs, policy, adapter, controls,
  documentation and the one-command workflow. It contains no generated public or
  evaluator artifacts.
• enterprise-authorization-opa-authzen-0.16.0-2-reference-run.zip
  (sha256:ef73d2fd92973843412a63f6fb34ecc1077db72e71630dc300d9127f2530e997):
  the source plus retained public artifacts, physically separate evaluator truth,
  raw AuthZEN requests/responses, sealed submissions, controls and reports.
• ASSET-METADATA.json records the correction, content-addressed source identity,
  versions, asset sizes, asset digests, result counts and unsupported claims.
• SHA256SUMS verifies every custom release asset.

The experiment source was produced outside a Git repository. It is therefore bound
by a deterministic source-file-set digest and final file manifest rather than a
source commit; this limitation is explicit in ASSET-METADATA.json.

Reproduce

sha256sum -c SHA256SUMS
unzip enterprise-authorization-opa-authzen-0.16.0-2-reproduction-kit.zip
cd enterprise-authorization-opa-authzen-0.16.0-2-reproduction-kit
./run.sh

First execution may require network access to obtain digest-pinned images and
hash-pinned packages. After the first run, ./reproduce.sh --offline performs the
complete two-run byte-identity check using the retained cache.

This experiment implements one combined RBAC/ABAC/ReBAC policy. Its negative
controls are intentionally broken variants, not competing authorization strategies.
Accountability is team-level rather than an explicit individual person-to-agent
assignment, and no SynthWorld HTML renderer was used.

Community experiments are owned and supported by their authors. They may be listed
as self-reported results in the SynthWorld
Experiment Results
Discussion category. A listing is not review, reproduction, endorsement or a
support commitment by SynthWorld maintainers.

Enterprise authorization OPA/AuthZEN reference experiment (SynthWorld 0.16.0)

Frozen enterprise authorization OPA/AuthZEN reference experiment

This is a frozen, unsupported record of agent-auth-3, an external consumer
experiment using idcognito-synthworld==0.16.0, Open Policy Agent 1.4.2 and an
experiment-owned adapter for OpenID Authorization API 1.0.

It is evidence of what was run. It is not a maintained adapter, AuthZEN
conformance claim, product certification, vendor comparison, production
enforcement result, or general authorization-correctness claim.

The experiment maps two fictional organization topologies using one deterministic
method, projects only physically separated public artifacts into OPA, binds every
AuthZEN request entity and evaluation-context field to its public benchmark cell,
seals submissions before scoring, and opens evaluator truth only in the offline
scorer capability.

Recorded results

• two topology lanes and 649 evaluation cells;
• 649/649 effective authorization decisions correct;
• 648/649 final decisions correct;
• 20/20 deliberately faulty policy and adapter controls detected;
• 16/16 fixed-cell-ID AuthZEN request mutations refused;
• 20/20 missing, modified, cross-topology and cross-version seal controls passed;
• 175/175 deterministic files byte-identical across two complete runs;
• fresh extraction of the reproduction ZIP completed successfully;
• every release and retained-reference checksum verified.

The residual binding errors are reported rather than hidden: canonical
account-to-principal binding is evaluator-only, so plausible same-kind substitutions
are not always decidable from public evidence.

Assets

• enterprise-authorization-opa-authzen-0.16.0-1-reproduction-kit.zip
  (sha256:3fa4b331cf82601d09169b50056d02ca711e805c61c4eed36a80db6bfc70e22b):
  source, pinned dependencies, two topology inputs, policy, adapter, controls,
  documentation and the one-command workflow. It contains no generated public or
  evaluator artifacts.
• enterprise-authorization-opa-authzen-0.16.0-1-reference-run.zip
  (sha256:ae4a5f78de9658c5411aa1cb68a91b764569d8ee5e7ed7e72068b41ee1ce69ee):
  the source plus retained public artifacts, physically separate evaluator truth,
  raw AuthZEN requests/responses, sealed submissions, controls and reports.
• ASSET-METADATA.json records the content-addressed source identity, versions,
  asset sizes, asset digests, result counts and unsupported claims.
• SHA256SUMS verifies every custom release asset.

The experiment source was produced outside a Git repository. It is therefore bound
by a deterministic source-file-set digest and the final file manifest rather than a
source commit; this limitation is explicit in ASSET-METADATA.json.

Reproduce

sha256sum -c SHA256SUMS
unzip enterprise-authorization-opa-authzen-0.16.0-1-reproduction-kit.zip
cd enterprise-authorization-opa-authzen-0.16.0-1-reproduction-kit
./run.sh

First execution may require network access to obtain digest-pinned images and
hash-pinned packages. After the first run, ./reproduce.sh --offline performs the
complete two-run byte-identity check using the retained cache.

This experiment implements one combined RBAC/ABAC/ReBAC policy. Its negative
controls are intentionally broken variants, not competing authorization strategies.
Accountability is team-level rather than an explicit individual person-to-agent
assignment, and no SynthWorld HTML renderer was used.

Community experiments are owned and supported by their authors. They may be listed
as self-reported results in the SynthWorld
Experiment Results
Discussion category. A listing is not review, reproduction, endorsement or a
support commitment by SynthWorld maintainers.

SynthWorld 0.15.0

Added

• Generated enterprise-agentic smoke vertical (#27). A new independently
  versioned configuration and Python API deterministically generates a bounded
  fictional organisation with humans, logical agents, runtimes, opaque
  credentials, resources, ownership, attenuated delegation, revocation,
  attribution, and authorised/adversarial action cases. Construction routes
  through the hardened base agentic projection, emits checksum-bound separate
  public/evaluator trees, derives denominated topology and integrity metrics, and
  can be selected explicitly with generate-enterprise-agentic --profile generated. Public-only and complete artifact-root loaders now reject inventory,
  canonicalization, checksum, cross-binding, semantic, derived-metric, and declared
  generator drift; dedicated CLI tasks validate and score external generated traces.
  An isolated-wheel consumer test and event-replay guidance preserve the oracle
  boundary. Existing fixed enterprise smoke and frozen Asteria/C08 bytes are
  unchanged; standard and longitudinal generated tiers remain follow-up work.
• Explorer v0.1 public projection contracts. A preview Python API projects
  the published Asteria Agentic v1 public package into deterministic graph and
  timeline records, with independently versioned evaluator overlays and layout
  manifests. Digest binding, UTC ordering, acyclic compound-node validation,
  collision-safe UUID5 identities, typed collection properties, and explicit
  public/evaluator separation are enforced. Interactive Cytoscape/ELK rendering,
  CLI integration, additional benchmark adapters, and generated-scale support
  remain deferred.

What's Changed

• Deploy Blume documentation to GitHub Pages by @bluntmachetti in #122
• Fix Blume heading duplication and guard docs migration by @bluntmachetti in #127
• Fix release README validation for string-form PEP 621 config by @bluntmachetti in #128
• Complete documentation front-door migration by @bluntmachetti in #129
• Move detailed guidance into canonical docs by @bluntmachetti in #130
• feat: define Explorer v0.1 projection contracts by @bluntmachetti in #126
• Document Enterprise Identity planning journeys by @bluntmachetti in #125
• Convert legacy guide to compatibility index by @bluntmachetti in #131
• Authorize bounded Hugging Face v0.14.0 dry run by @bluntmachetti in #123
• Add generated enterprise-agentic smoke vertical by @bluntmachetti in #132
• Brand the SynthWorld documentation shell by @bluntmachetti in #133
• Complete generated enterprise-agentic consumer path by @bluntmachetti in #134
• Prepare SynthWorld 0.15.0 by @bluntmachetti in #135

Full Changelog: v0.14.0...v0.15.0

SynthWorld 0.14.0

Added

• Independent C08 evidence-binding v2 contracts. Asteria and enterprise
  lineages now have separately versioned public input, evaluator truth,
  submission, report, manifest, deterministic reference-generation, and
  evidence-quality metric contracts. Public inputs expose opaque binding
  handles and same-kind distractors while exact required bindings remain in
  evaluator truth. Existing v1 contracts and frozen bytes remain unchanged.

• C08 v2 candidate registry metadata. Repository-local metadata records the
  Asteria and enterprise C08 v2 benchmark identities as candidates with pending
  publication gates. These entries do not publish either benchmark externally
  or claim external hosting, viewer support, download availability, or
  re-download verification.

• Independent C08 v2 frozen-artifact candidates. Asteria now commits an exact
  five-file root/public/evaluator manifest tree; enterprise commits an exact
  four-file root-manifest/SHA256SUMS tree with a packaged fail-closed loader and
  fixed-seed identity comparison. Both public contracts use opaque binding
  handles plus same-kind distractors instead of exposing a unique kind-to-ID
  answer. Independent manifest schemas, expanded v1 hash locks, explicit offline
  report scope, and exactly two metric-only discrimination records accompany the
  candidate bytes. Native adversarial findings and repository verification gates
  are recorded as resolved. Nothing is externally published or deployed.

• A curated benchmark registry now records independent lifecycle, benchmark-kind,
  evaluation-mode, artifact-sensitivity, integrity, and publication-gate evidence
  for every current benchmark family.

• A source-complete Blume documentation site. The repository now carries a
  structured public documentation tree, generated capability and benchmark
  catalogues, dark-preview builds, documentation-impact routing, and distribution
  audits. The site is ready for a separately authorized deployment but is not
  published by this release.

• Guarded Hugging Face publication planning. A local publication manifest,
  offline validator, and protected dry-run workflow bind publication authority to
  the resolved benchmark registry. Uploads remain disabled: no benchmark or
  operation is authorized and the outstanding external gates remain explicit.

• A safely fictional EADS-shaped adapter example. The repository-only,
  humans-only example accepts bounded JSON or YAML fixtures, emits separately
  typed public and evaluator projections, and records deterministic path-bound
  provenance. It is not shipped in the wheel, performs no network access, and
  makes no real-EADS compatibility claim.

• Future enterprise authority designs. Reviewed design records reserve
  non-conflicting C15/C16 v1 and Face B contract, schema, package, benchmark, and
  manifest identities. They define canonical normalization, collision rejection,
  UUID5 domain separation, public/evaluator boundaries, and staged gates without
  implementing or freezing those deferred families.

Changed

• Agentic evidence metrics and deployment-pattern coverage now state their
  reporting-only limits, metric polarity, denominators, null and abstention
  semantics, and declaration-versus-observation boundary without changing metric
  values, scoring versions, schemas, or frozen artifacts.
• Benchmark, documentation, release, and Hugging Face governance now fail closed
  on same-version frozen-inventory additions, compare push transitions with the
  pre-push commit, audit the exact documentation distribution, and publish the
  already-verified release artifact rather than rebuilding it.

What's Changed

• Clarify agentic evidence metrics and pattern declarations by @bluntmachetti in #100
• feat(agentic): add C08 evidence-binding v2 contracts by @bluntmachetti in #112
• feat(agentic): freeze C08 v2 benchmark artifacts by @bluntmachetti in #113
• Register C08 v2 benchmark candidates by @bluntmachetti in #114
• Add governed capability inventory by @bluntmachetti in #103
• feat: govern benchmark publication registry by @bluntmachetti in #104
• docs: add stable journey-led documentation by @bluntmachetti in #107
• docs: add hermetic Blume dark preview by @bluntmachetti in #108
• docs: integrate registry-backed catalogue by @bluntmachetti in #110
• Add guarded Hugging Face publication controls by @bluntmachetti in #109
• Reconcile agentic governance evidence claims by @bluntmachetti in #115
• Reconcile Blume publication governance by @bluntmachetti in #116
• Reconcile C15 C16 contract design by @bluntmachetti in #117
• Rebuild fictional EADS shaped adapter by @bluntmachetti in #118
• Reconcile Face B compiled universe design by @bluntmachetti in #119
• Prepare SynthWorld 0.14.0 by @bluntmachetti in #120
• Support core metadata 2.5 publication by @bluntmachetti in #121

Full Changelog: v0.13.0...v0.14.0

SynthWorld 0.13.0

Added

• A deterministic enterprise identity and access surface, with authorization
  evaluated against standards-shaped projections (#7, #27). SynthWorld now
  generates a fixed enterprise identity/access universe — organisations, units,
  populations, principals, accounts, groups, roles, permissions, entitlements,
  and opaque authorization targets — and evaluates access against it through
  three bounded oracles: a directory RBAC oracle with role and group closure,
  a bounded ABAC oracle over declared attributes, and a bounded ReBAC
  oracle over relationship tuples. Access derivation is an implementation detail
  of the oracle rather than a published topology, so no artifact describes the
  model as an identity topology.
  New modules: enterprise, enterprise/rbac, enterprise/abac,
  enterprise/rebac, enterprise/authorization, enterprise/conformance,
  enterprise/identity_fabric, agentic/enterprise.
• Vendor-neutral projections so a real authorization engine can consume the
  world without a bespoke adapter. enterprise/projections/ emits an RFC
  7643-style SCIM user/group projection, an OpenFGA authorization model
  and relationship tuples for the bounded ReBAC subset, and an AuthZEN 1.0
  request projection with per-field provenance. Each projection carries an
  explicit mapping profile recording what it can and cannot represent, so a
  projection gap is declared rather than silently lossy. A Shared Signals/CAEP
  mapping profile is declared, but temporal event emission is deliberately
  deferred and no SET envelope is constructed yet.
• Two smoke benchmarks and a graded assurance ladder. An enterprise identity
  fabric smoke benchmark and an enterprise agentic smoke benchmark publish public
  inputs and evaluator truth under enterprise-identity-access-contract/; a
  contextual access benchmark profile publishes under
  contextual-access-contract/. continuous_assurance adds smoke,
  standard, longitudinal, and held_out tiers governing assurance cadence.
  Note these are assurance tiers, not generated-world scale tiers — the
  enterprise_agentic scale ladder #27 asks for remains open at smoke only.
• An executable agent-authority run protocol. agent_authority and
  assurance add a staged run protocol with signed execution receipts, component
  provenance, and evidence claims, so an evaluation run is reconstructible from
  its receipt rather than trusted on assertion.
• The ambiguity v2 pack gets its difficulty from a computed error floor, not a
  codebook (#80). The v1-style surfaces encoded each identity index in cleartext, so a
  ~30-line normaliser recovered every relation and scored 1.0000; two successor designs
  fell to pool inversion. The fix follows the reviewed plan: each kind draws a base from
  a pool of confusable clusters (Sorensen/Sorenson/Soerensen, a transposed
  phone, a swapped day/month), EQUAL/NEAR share the base while FAR redraws from a
  stationary mixture, and every value passes through one structured-noise operator
  applied identically under every relation. Identity recovery stays free and expected;
  the relation is carried by overlapping distance distributions. The pack publishes its
  genie floor — the Bayes error of the generator, estimated with a stated N and
  Wilson interval and keyed to a digest of every decision-relevant constant — plus the
  enumerated channel invariants (kernel stationarity, an identical one-value marginal
  under every relation, a per-base sibling-landing mass gate, form bijectivity and
  constant cross-form distance, and an artifact-factorization check) and a gated
  technique premium, so real resolution technique is rewarded rather than anti-taught.
  New modules: ambiguity_evidence, ambiguity_surfaces, ambiguity_channel,
  ambiguity_floor, with v2 serialization/metrics/baselines support and
  examples/compute_ambiguity_floor.py.
• Agent-authority and contextual-access receipt builders now seal honest
  failed-run receipts: a failed product execution produces a manifest with
  execution_status=failed and evaluation_status=not_evaluated that binds
  only the product-stage artifacts and never loads evaluator truth, so an
  assurance corpus can no longer be structurally biased toward successful runs.
  Receipt validation enforces the paired statuses and the product-only artifact
  inventory, and run manifests whose evidence claim is not supported by the
  systems under test (live-lab claims with reference-only components) are
  rejected. Managed-service provenance in not_exposed observability states
  additionally forbids evidence references, and the contextual execution
  receipt leaves stimulus_digest unset because that lineage executes the
  public input directly.
• Contextual-access receipts now expose the same explicit two-phase live-run
  boundary as agent-authority receipts. External runners may finish and attribute
  the product stage before constructing completion metadata; the finalizer then
  replays the plan, public input, adapter, component inventory, provenance, and
  artifact digests before evaluator truth is loaded. Existing deterministic
  contextual receipt bytes remain unchanged.
• An opt-in disposable agent-authority reference deployment now executes the
  public enterprise-agentic smoke world across isolated Docker networks. It
  produces live observation-v2 receipts covering L01-L06, the exact declared
  L07 baseline/SUT inventory, and measured/unsupported L08 targets, while
  keeping runtime credentials in a destroyed named volume and scanning canary
  and token markers out of receipts, logs, and container metadata. A new
  two-phase receipt finalizer lets live runners record completion metadata only
  after external execution without exposing evaluator truth before product
  output is durably staged.
• Agent-authority observation schema 2.0.0 corrects L06 clock semantics without
  changing the frozen observation-v1 schema. It records one explicit monotonic
  revocation epoch, non-negative acknowledgement offsets, and signed send/completion
  offsets so pre-revocation in-flight requests are representable. Receipt validation
  dispatches v1/v2 observations and binds them to scoring formulas 1.0.0/2.0.0;
  migration guidance forbids guessing offsets from ambiguous v1 rows.
• The 12-case authority-change governance conformance fixture from #73 is now an
  additive frozen benchmark. Its public and evaluator payloads remain physically
  separate; their visibility manifests and exact raw bytes are path-bound by a
  packaged SHA256SUMS, verified by the packaged loader API, regeneration tests,
  and isolated-wheel checks. No existing golden bytes changed.
• The broker-removal pack is scored through the unified evaluator: evaluate_broker_removal
  projects each family's headline ratio into the standard EvaluationReport, the CLI gains
  synthworld evaluate broker, and examples/evaluate_broker_adapter.py is the worked
  Idcognito-style adapter #5 asked for - public timeline in, versioned assessment out,
  scored against regenerated truth. Closes the last acceptance criteria of #5.
• Propagation lag is representable and scored (#65). Downstream copies carry their own
  removal tick (None never goes), a new slow_propagation case has copies that catch up
  late rather than never, a submission can predict the completion tick, and
  propagation_lag reports mean absolute error with support. The credulous baseline now
  predicts completion at confirmation - "done means done everywhere" - and eats a 14-tick
  error on exactly the case built to price that claim; the example adapter's modest grace
  period cuts it to 4.

Changed

• Ambiguity grammar 2.0.0, v2 schema 2.1.0. render_relation/render_value
  delegate to the structured-noise channel; the old _SPACE/_surface codebook is gone.
  Relation.EQUAL no longer means "byte-identical" — it is one value transcribed once
  per record, rendered identically only with probability sigma — and the charter
  docstrings that claimed otherwise are rewritten. The v1 pack and its frozen artifacts
  are untouched and stay byte-identical.
• display_name renders family, given (#86). The two name kinds are scored as
  separate evidence, so the boundary between them must be readable off the value; the
  old given family lost it whenever a pool entry carried a space. No rendered name
  contains ", ", so the split is unambiguous.
• Temporal schema 1.2.0. ListingTruth.downstream_refs (bare strings) becomes
  downstream_copies with per-copy removal ticks, and a recorded reappearance must now
  coincide with a published LISTING_REAPPEARED event - truth that disagrees with the
  public timeline is refused as corrupt input. Deliberately asymmetric with removal, which
  stays unpinned because a confirmation is the broker's claim and the phantom case exists
  to show the claim can be false. BrokerAssessment moves to 1.1.0 for the new
  prediction field; propagation state is now read as of the assessed tick, so a slowly
  propagating deletion no longer scores identically to one that never propagates.
• DenominatedMetric moved to synthworld.models, below the evaluation/partition import
  cycle it was about to create; ambiguity_partition re-exports it unchanged.

What's Changed

• Close the broker pack: lag scored, evaluator unified, adapter worked (#5, #65) by @bluntmachetti in #95
• Give the ambiguity v2 pack a computed error floor, not a codebook (#80) by @bluntmachetti in #96
• Add the enterprise authorization testbed: identity fabric, contextual access, and governance benchmarks (PR1-PR13) by @bluntmachetti in #97
• Prepare SynthWorld 0.13.0: document the...

SynthWorld 0.12.0

Changed

• Breaking (ambiguity answer key): same_name_and_date_of_birth is now
  insufficient, not separate (#77). The pair is two people in canonical truth, but
  the public evidence — matching name and birth date, nothing distinguishing them —
  cannot justify concluding it. The pack's first consumer abstained on exactly this pair
  and independently gave the same reason. Membership truth is unchanged; the public and
  memberships artifacts are byte-identical, and only the dispositions artifact was
  re-cut (new digest in GOLDEN_REVIEW.md). A resolver scored against the old key that
  answered separate here was being rewarded for clairvoyance; one that abstains is now
  scored correctly.

• EVALUATION_SCHEMA_VERSION is 0.2.0. TaskMetric gains optional family and
  support_meaning, so every task's report carries two more keys - extraction,
  entity resolution, relationship inference and risk included, even though none of
  their metrics changed meaning. The wire shape is what moved, so the schema knob is
  what moves; no per-task scoring version changes, because a scoring version here means
  the metric definitions and those are untouched. A stored 0.1.0 report does not
  load under the new model — the report's schema_version is a single-value literal —
  so read archived reports with the library version that wrote them.

• Agentic metrics are grouped into five families and every denominator says what it
  counts, so a report can be read by family and each ratio re-derived rather than
  trusted. No metric value moves. The split carrying the most information is
  observability against the rest: an agent can decide well and record badly, or the
  reverse, and those need different fixes. Measured on the reference trace, wrecking
  the recording drops observability to 0.25 while identity resolution, authorization
  and delegation stay at 1.0; wrecking the decisions drops authorization to 0.40 while
  observability stays at 1.0.

• AGENTIC_BENCHMARK.md gains a per-metric glossary: what each measures, what 0.0 and
  1.0 mean, and its denominator. Sixteen of the twenty metrics had no mention in any
  top-level document, so the only way to learn what they measured was to read the
  scorer or diff scores between policies. (They were cited in
  agent-authority-contract/control-catalogue.yaml and its design-intent notes, which
  a first version of this entry overlooked while claiming a repository-wide count.)

What's Changed

• Correct the 0.11.0 release notes: nine of eleven, not eleven by @bluntmachetti in #70
• Give agentic metrics families, denominators, and a glossary (#71) by @bluntmachetti in #72
• Create the GitHub release from the tag, not by hand by @bluntmachetti in #74
• Derive ambiguity dispositions from evidence instead of hand-authoring them (#62) by @bluntmachetti in #75
• Stop disposition_of concluding things a resolver would not (#81, #85) by @bluntmachetti in #87
• Stop the third disposition class starving by @bluntmachetti in #88
• Stop rewarding clairvoyance on same_name_and_date_of_birth (#77) by @bluntmachetti in #89
• Tranche closeout: honest v2 framing, v1 result sizing, key custody by @bluntmachetti in #93
• Prepare SynthWorld 0.12.0 by @bluntmachetti in #94

Full Changelog: v0.11.0...v0.12.0

SynthWorld 0.11.0

Published on PyPI as idcognito-synthworld==0.11.0 (wheel and sdist). The tag is annotated and SSH-signed; git verify-tag v0.11.0 checks it without trusting the forge.

Read this before scoring against an ambiguity pack

Nine of eleven known channels through which the ambiguity pack's answer key was recoverable from its public artifact are closed in this release. Two remain open — see #68.

If you generated evaluation packs with an earlier version, regenerate them. A system under test could reach the right answer without doing the task, so scores measured against those packs do not mean what they appear to. Regenerating with 0.11.0 does not make a pack safe against the two remaining channels; it does close the other nine, which is a large difference.

The frozen canonical pack was affected and is re-cut here, with new digests recorded in GOLDEN_REVIEW.md. Every other frozen benchmark — Asteria Agentic v1 included — is byte-identical.

The one worth understanding

Eight of the nine were metadata bound to the label: collection ordering, name pools indexed by a scenario ordinal, positional record identifiers, source types constant per scenario, repetition counts, attribute counts, a locality token, cross-listing multiplicity.

The ninth was different in kind. The substitution plan was a deterministic function of a published seed over canonical values that live in public source, so it could be recomputed and inverted rather than correlated — 0.929 disposition recovery against a 0.467 baseline, reading no identity evidence. No statistical leak detector could have found it, because there is no structure in the emitted values to detect.

generate_ambiguity_variant therefore now requires a key that is never serialized:

from synthworld.ambiguity_variants import UNKEYED, generate_ambiguity_variant

generate_ambiguity_variant(seed=42, key=UNKEYED)          # reproduces published packs
generate_ambiguity_variant(seed=42, key=secrets.token_bytes(16))  # for evaluation

A held-out seed protects surface values and nothing else — it is published inside the artifact. A held-out key protects the artifact.

New task families

Human identity-resolution ambiguity pack; oracle-free provider-shaped search projection with scoring; deterministic temporal identity worlds; broker deletion-and-reappearance scoring; consumer-neutral run receipts; a households profile; and a recoverability detector.

Contract versions

BROKER_SCORING_VERSION 2.0.0 and TEMPORAL_SCHEMA_VERSION 1.1.0 — both on schemas that have never shipped in a package, so nothing released changes meaning. Data contracts are versioned independently of the package; see DATA_DICTIONARY.md.

What's Changed

• Publish Asteria trace conventions for chains, evidence, and side effects by @bluntmachetti in #35
• Feature agent authority in README by @bluntmachetti in #37
• Agent authority contract package and validate agentic-trace by @bluntmachetti in #38
• Revise planned work items in README by @bluntmachetti in #39
• Harden the shipped agentic surface: derive expected_policy_version, close the case-kind vocabulary by @bluntmachetti in #40
• Measure recoverable truth, and add the households_and_workplaces profile by @bluntmachetti in #44
• Document the core identity world as a smoke surface by @bluntmachetti in #45
• Sign release tags, and document why by @bluntmachetti in #46
• Emit ground truth the evaluator accepts, and fix reproducibility by @bluntmachetti in #47
• Complete the households profile: locality, realism metrics, manifest, CLI by @bluntmachetti in #48
• Freeze a households smoke fixture and document generation cost by @bluntmachetti in #49
• Add the human identity-resolution ambiguity pack by @bluntmachetti in #50
• Add an oracle-free, provider-shaped search projection by @bluntmachetti in #51
• Protect the public consumer-integration boundary by @bluntmachetti in #56
• Fix semantically corrupt ambiguity variants by @bluntmachetti in #54
• Preserve complete ambiguity partition evaluation by @bluntmachetti in #57
• Add consumer-neutral reproducible run receipts by @bluntmachetti in #58
• Close three oracle channels in the ambiguity pack by @bluntmachetti in #59
• Score the search projection, and document what it is by @bluntmachetti in #53
• Set out security scope and route reports from CONTRIBUTING by @bluntmachetti in #55
• Add the privacy-exposure temporal slice (#2) by @bluntmachetti in #60
• Score broker deletion and reappearance (#5) by @bluntmachetti in #61
• Give attribution evidence, refuse an impossible claim, bump the contracts by @bluntmachetti in #66
• Key the substitution plan: a public seed is not a secret by @bluntmachetti in #67
• Prepare SynthWorld 0.11.0 by @bluntmachetti in #69

Full Changelog: v0.10.0...v0.11.0