Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
36 commits
Select commit Hold shift + click to select a range
8ef501f
feat(orchestration): add scoped repair reviews
Ldsystem Sep 5, 2026
5265c96
fix(orchestration): bind task repair acceptance
Ldsystem Sep 5, 2026
09982b1
fix(orchestration): verify task repair provenance
Ldsystem Sep 5, 2026
cfa75a3
feat(orchestration): require subagent task ownership
Ldsystem Sep 5, 2026
53bb36c
fix(orchestration): advance serialized task waves
Ldsystem Sep 5, 2026
11cd540
feat(orchestration): bound compiled context projection
Ldsystem Sep 5, 2026
1cbe41c
test(orchestration): add WOR-108 closure fixtures
Ldsystem Sep 5, 2026
39eab9c
fix(evals): type WOR-108 fixture identities
Ldsystem Sep 5, 2026
02ebc7f
test(orchestration): prove bounded repair frontier
Ldsystem Sep 5, 2026
473e4b2
fix(orchestration): exercise production ownership scheduler
Ldsystem Sep 5, 2026
a03b96f
test(orchestration): close WOR-108 SG oracle gaps
Ldsystem Sep 5, 2026
58c3f23
fix(orchestration): enforce task ownership at acceptance
Ldsystem Sep 5, 2026
fa663f0
fix(orchestration): keep mutation events runtime-only
Ldsystem Sep 5, 2026
cc19c68
fix(orchestration): admit exact dependency deltas
Ldsystem Sep 5, 2026
39fff06
fix(orchestration): enforce repair runtime ownership
Ldsystem Sep 5, 2026
5d701f3
test(orchestration): isolate dispatcher module loading
Ldsystem Sep 5, 2026
084e339
fix(orchestration): compose accepted repair chains
Ldsystem Sep 5, 2026
37bd559
fix(orchestration): anchor accepted repair chains
Ldsystem Sep 5, 2026
3529b17
fix(orchestration): bind dependency repairs to plan
Ldsystem Sep 5, 2026
9f46c81
fix(orchestration): validate cumulative accepted result deltas
Ldsystem Sep 5, 2026
6203ff1
fix(orchestration): enforce cumulative review succession
Ldsystem Sep 6, 2026
77c525a
feat(orchestration): add accepted-result recovery receipts
Ldsystem Sep 6, 2026
d0e42c2
fix(orchestration): scope recovery lookup to expected base
Ldsystem Sep 6, 2026
cd162e5
fix(orchestration): scope recovery receipts to proposals
Ldsystem Sep 6, 2026
1e68077
fix(orchestration): adopt published recovery results
Ldsystem Sep 6, 2026
5a483de
fix(orchestration): reference oversized review diffs
Ldsystem Sep 6, 2026
ff6b73a
chore(orchestration): converge accepted WOR-108 results
Ldsystem Sep 6, 2026
ff96d97
fix(evals): enforce migration impact completeness
Ldsystem Sep 6, 2026
59699f8
fix(evals): bind exact WOR-108 production oracles
Ldsystem Sep 6, 2026
f221a61
fix(orchestration): close task acceptance evidence gaps
Ldsystem Sep 6, 2026
89537df
test(wor105): bind transition to accepted release
Ldsystem Sep 6, 2026
b315067
fix(orchestration): close acceptance review gaps
Ldsystem Sep 6, 2026
ea0645b
test(wor108): include repaired WOR105 evidence surfaces
Ldsystem Sep 6, 2026
aff7f7f
fix(orchestration): normalize mutation evidence paths
Ldsystem Sep 6, 2026
f1d66e0
fix(orchestration): require archive ownership evidence
Ldsystem Sep 6, 2026
fdd5940
test(orchestration): refresh WOR-108 migration impact
Ldsystem Sep 6, 2026
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions evals/wor105/components/native-transition-record.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,11 @@ schema: wor105-native-transition-v1
issue: WOR-105
transition_task: task-b06r
enforcement_transition: bootstrap_policy_to_native
source_identity: 10aa3aceeea4c440feaccb484b897af67cd0df0b+repository-evidence-sha256:279f142fad6ea3005370b9eaa82d7c88444032e668309173e9230754b28c40ae
source_identity: 9dce5df221485174d6179f713e8b179bbc20567a+repository-evidence-sha256:279f142fad6ea3005370b9eaa82d7c88444032e668309173e9230754b28c40ae
review_path: .work-bundle/orchestration/reviews/WOR-105-task-b06r-kernel-review-accepted.yaml
review_sha256: d638b8959b4db57dd33e072b01428d6ce89f65a9771c05eff26d8e40d4293ebd
accepted_commit: 10aa3aceeea4c440feaccb484b897af67cd0df0b
accepted_tree: 3f70163e3aef240b0f1207f43433874c9139a7fb
accepted_commit: 9dce5df221485174d6179f713e8b179bbc20567a
accepted_tree: 5a1f38355eae8068bab528923e807ce54e6f6fe5
integrated_validation: {id: VAL-B06R-TEST, result: passed, tests: 300}
handoff_validation: {id: VAL-B06R-IDENTITY, result: passed, adversarial_cases: 5}
participant_handoffs:
Expand Down
2 changes: 1 addition & 1 deletion evals/wor105/freeze-manifest.json
Original file line number Diff line number Diff line change
Expand Up @@ -72,7 +72,7 @@
"components": {
"profile": {
"path": "evals/wor105/components/native-transition-record.yaml",
"sha256": "aa01163346022ddc9b55481cf1698614da08a1c0f75a98d38b5ce3008b57cd56"
"sha256": "0d39d8709d8b4884a93384a51829d147e465dbf674e2c122d324d67c1283205a"
},
"fixtures": {
"path": "evals/wor105/fixtures",
Expand Down
24 changes: 12 additions & 12 deletions evals/wor105/results.jsonl

Large diffs are not rendered by default.

38 changes: 38 additions & 0 deletions evals/wor108/contracts-v1.schema.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,38 @@
{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"$id": "urn:work-bundle:wor108:closure-contracts:v1",
"title": "WOR-108 legacy closure fixture registry",
"type": "object",
"additionalProperties": false,
"required": ["contract", "evaluation_id", "fixtures"],
"properties": {
"contract": {"const": "wor108-closure-fixtures-v1"},
"evaluation_id": {"const": "wor108-legacy-closure-v1"},
"fixtures": {
"type": "array",
"minItems": 22,
"maxItems": 22,
"items": {"$ref": "#/$defs/fixture"}
}
},
"$defs": {
"fixture": {
"type": "object",
"additionalProperties": false,
"required": ["fixture_id", "area", "title", "oracle", "pytest_nodes", "expected"],
"properties": {
"fixture_id": {"type": "string", "pattern": "^(RF|SG|CTX)-[0-9]{2}$"},
"area": {"enum": ["repair_frontier", "subagent_ownership", "context_projection"]},
"title": {"type": "string", "minLength": 1},
"oracle": {"const": "pytest"},
"pytest_nodes": {
"type": "array",
"minItems": 1,
"uniqueItems": true,
"items": {"type": "string", "pattern": "^tests/test_wor108_[a-z_]+\\.py::test_[a-z0-9_]+$"}
},
"expected": {"const": "passed"}
}
}
}
}
198 changes: 198 additions & 0 deletions evals/wor108/fixtures.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,198 @@
{
"contract": "wor108-closure-fixtures-v1",
"evaluation_id": "wor108-legacy-closure-v1",
"fixtures": [
{
"fixture_id": "RF-01",
"area": "repair_frontier",
"title": "Initial and repair review modes are native closed contract values",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_review_frontier.py::test_rf_01_h1_repairs_recorded_a_without_reacquiring_unrecorded_latent_b"],
"expected": "passed"
},
{
"fixture_id": "RF-02",
"area": "repair_frontier",
"title": "Repair frontier binds prior blocking findings, frozen evidence, and exact identities",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_review_frontier.py::test_rf_02_repair_frontier_binds_prior_findings_evidence_and_exact_identities"],
"expected": "passed"
},
{
"fixture_id": "RF-03",
"area": "repair_frontier",
"title": "Repair reuses frozen evidence by reference without copying whole review history",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_review_frontier.py::test_rf_03_capable_untouched_invariant_stays_blocking_during_narrow_repair"],
"expected": "passed"
},
{
"fixture_id": "RF-04",
"area": "repair_frontier",
"title": "Material redesign or authority boundary change forces a fresh initial review",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_review_frontier.py::test_rf_04_material_change_requires_fresh_initial_review"],
"expected": "passed"
},
{
"fixture_id": "RF-05",
"area": "repair_frontier",
"title": "Fresh initial review requires a fresh capable independent reviewer identity",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_review_frontier.py::test_rf_05_reset_rejects_reused_repair_reviewer_identity"],
"expected": "passed"
},
{
"fixture_id": "RF-06",
"area": "repair_frontier",
"title": "Repair rejects stale or relabelled repaired identities",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_review_frontier.py::test_rf_06_final_broad_integrated_review_rediscovers_and_classifies_latent_b"],
"expected": "passed"
},
{
"fixture_id": "RF-07",
"area": "repair_frontier",
"title": "Repair catches affected-boundary regressions and retains the original execution binding baseline",
"oracle": "pytest",
"pytest_nodes": [
"tests/test_wor108_review_frontier.py::test_rf_07_repair_still_detects_regression_in_affected_boundary",
"tests/test_wor108_context_projection.py::test_rf_07_brief_rebuild_retains_original_execution_binding_and_baseline"
],
"expected": "passed"
},
{
"fixture_id": "RF-08",
"area": "repair_frontier",
"title": "Repair packages stay bounded for task and stage targets",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_review_frontier.py::test_rf_08_repair_packet_is_bounded_for_task_and_stage_targets"],
"expected": "passed"
},
{
"fixture_id": "SG-01",
"area": "subagent_ownership",
"title": "Execute-plan selection implicitly requires subagent task ownership",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_subagent_ownership.py::test_sg01_execute_plan_requires_implicit_subagent_ownership"],
"expected": "passed"
},
{
"fixture_id": "SG-02",
"area": "subagent_ownership",
"title": "No-subagent execution fails closed before task mutation",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_subagent_ownership.py::test_sg02_no_subagent_fails_closed_before_mutation"],
"expected": "passed"
},
{
"fixture_id": "SG-03",
"area": "subagent_ownership",
"title": "Legacy prefer_subagent metadata has no behavioral effect",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_subagent_ownership.py::test_sg03_legacy_preference_has_no_behavioral_effect"],
"expected": "passed"
},
{
"fixture_id": "SG-04",
"area": "subagent_ownership",
"title": "Controller mutation of task write scope is rejected despite green validation",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_context_projection.py::test_completed_task_acceptance_rejects_controller_task_scope_mutation"],
"expected": "passed"
},
{
"fixture_id": "SG-05",
"area": "subagent_ownership",
"title": "Independent disjoint tasks fan out before any wait",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_subagent_ownership.py::test_sg05_independent_disjoint_tasks_dispatch_before_wait_and_convergence"],
"expected": "passed"
},
{
"fixture_id": "SG-06",
"area": "subagent_ownership",
"title": "Dependent, overlapping, and same-workspace tasks remain serialized and make progress",
"oracle": "pytest",
"pytest_nodes": [
"tests/test_wor108_subagent_ownership.py::test_sg06_dependent_or_overlapping_tasks_are_serialized",
"tests/test_wor108_subagent_ownership.py::test_sg06_same_execution_workspace_is_never_fanned_out",
"tests/test_wor108_subagent_ownership.py::test_sg06_serialized_tasks_progress_across_successive_waves"
],
"expected": "passed"
},
{
"fixture_id": "SG-07",
"area": "subagent_ownership",
"title": "Repair mutation remains subagent-owned",
"oracle": "pytest",
"pytest_nodes": [
"tests/test_wor108_subagent_ownership.py::test_sg07_repair_retains_owner_binding_baseline_evidence_and_frontier",
"tests/test_wor108_subagent_ownership.py::test_sg07_repair_allows_only_an_authorized_replacement_owner"
],
"expected": "passed"
},
{
"fixture_id": "SG-08",
"area": "subagent_ownership",
"title": "Ownership provenance is provider-neutral and rejects visibility-specific fields",
"oracle": "pytest",
"pytest_nodes": [
"tests/test_wor108_subagent_ownership.py::test_sg08_native_mechanisms_preserve_task_binding_handoff_and_validation_semantics",
"tests/test_wor108_subagent_ownership.py::test_visibility_specific_provenance_is_rejected"
],
"expected": "passed"
},
{
"fixture_id": "CTX-01",
"area": "context_projection",
"title": "Unrelated provenance growth does not inflate briefs or repair packages",
"oracle": "pytest",
"pytest_nodes": [
"tests/test_wor108_context_projection.py::test_ctx_01_unrelated_runtime_history_does_not_inflate_unchanged_task_brief",
"tests/test_wor108_context_projection.py::test_ctx_01_repair_package_uses_frontier_without_reacquiring_review_history"
],
"expected": "passed"
},
{
"fixture_id": "CTX-02",
"area": "context_projection",
"title": "Semantic authority is retained once and referenced by stable ID elsewhere",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_context_projection.py::test_ctx_02_semantic_authority_is_retained_once_and_referenced_by_id"],
"expected": "passed"
},
{
"fixture_id": "CTX-03",
"area": "context_projection",
"title": "Executor capability projection contains only allocated capabilities and reasons",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_context_projection.py::test_ctx_03_executor_capability_projection_contains_only_allocated_capabilities"],
"expected": "passed"
},
{
"fixture_id": "CTX-04",
"area": "context_projection",
"title": "Successful evidence projects compact digest and freshness receipts without history or stdout",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_context_projection.py::test_ctx_04_success_evidence_projects_compact_receipt_not_history_or_stdout"],
"expected": "passed"
},
{
"fixture_id": "CTX-05",
"area": "context_projection",
"title": "Failed evidence expands only for an allowed explicit reason",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_context_projection.py::test_ctx_05_failed_evidence_expands_only_with_allowed_reason"],
"expected": "passed"
},
{
"fixture_id": "CTX-06",
"area": "context_projection",
"title": "Compiled context has telemetry, lazy histories, no retrieval escape hatch, and no arbitrary hard limit",
"oracle": "pytest",
"pytest_nodes": ["tests/test_wor108_context_projection.py::test_ctx_06_no_retrieval_escape_hatch_and_context_metrics_use_existing_telemetry"],
"expected": "passed"
}
]
}
73 changes: 73 additions & 0 deletions evals/wor108/migration-impact.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,73 @@
{
"contract": "wor108-migration-impact-v1",
"issue": "WOR-108",
"evaluation_id": "wor108-legacy-closure-v1",
"accepted_legacy_baseline": {
"commit": "9dce5df221485174d6179f713e8b179bbc20567a",
"tree": "5a1f38355eae8068bab528923e807ce54e6f6fe5"
},
"changed_surfaces": {
"public_contracts": [
"references/assets/orchestration/contract/handoff-executor-result-v1.md",
"references/assets/orchestration/contract/stage-review-v1.schema.json",
"references/assets/orchestration/contract/task-v1.md",
"references/assets/orchestration/workflow.md"
],
"instructions": [
"references/assets/template/AGENTS.md",
"references/assets/template/bootstrap.yaml",
"references/assets/template/project.yaml",
"rules/orchestration/orch-handoff-required.md",
"rules/work-bundle/wb-project-context-preflight.md",
"skills/orch-create-handoff/SKILL.md",
"skills/orch-doctor/SKILL.md",
"skills/orch-execute-plan/SKILL.md",
"skills/wb-initialize-project/SKILL.md"
],
"operations": [
"scripts/orchestration/dispatcher.py",
"scripts/orchestration/execution_context.py",
"scripts/orchestration/plans.py",
"scripts/orchestration/review_runtime.py",
"scripts/orchestration/task_ownership.py",
"scripts/work-bundle/control_plane.py",
"scripts/work-bundle/core.py",
"scripts/work-bundle/dispatcher.py",
"scripts/work-bundle/migration.py",
"scripts/work-bundle/project.py",
"scripts/work-bundle/reviewer_workspace.py",
"scripts/work-bundle/stage_events.py"
],
"fixtures_and_evaluations": [
"evals/wor105/components/native-transition-record.yaml",
"evals/wor105/freeze-manifest.json",
"evals/wor105/results.jsonl",
"evals/wor108/contracts-v1.schema.json",
"evals/wor108/fixtures.json",
"evals/wor108/migration-impact.json",
"evals/wor108/verify.py",
"references/evals/orchestration/evals.json",
"tests/test_control_plane_v4.py",
"tests/test_multi_repository_member.py",
"tests/test_orchestration_execution_context.py",
"tests/test_orchestration_skill_rule_boundary.py",
"tests/test_orchestration_workflow_contracts.py",
"tests/test_project_initialization.py",
"tests/test_registry_layout_migration.py",
"tests/test_wor108_closure.py",
"tests/test_wor108_context_projection.py",
"tests/test_wor108_review_frontier.py",
"tests/test_wor108_subagent_ownership.py"
]
},
"evaluation_identities": [
"RF-01", "RF-02", "RF-03", "RF-04", "RF-05", "RF-06", "RF-07", "RF-08",
"SG-01", "SG-02", "SG-03", "SG-04", "SG-05", "SG-06", "SG-07", "SG-08",
"CTX-01", "CTX-02", "CTX-03", "CTX-04", "CTX-05", "CTX-06"
],
"handoff_constraints": {
"final_git_identity": "post_acceptance_handoff_only",
"advance_wor107": false,
"touch_work_bundle_mcp": false
}
}
Loading