@velum-labs/routekit-eval-setup 1.3.2 → 1.5.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/dist/adapters/authoring-responses-request.d.ts +5 -0
- package/dist/adapters/authoring-responses-request.js +38 -0
- package/dist/adapters/evaluation-evidence-freshness.d.ts +7 -0
- package/dist/adapters/evaluation-evidence-freshness.js +76 -0
- package/dist/adapters/git-task-history.d.ts +67 -0
- package/dist/adapters/git-task-history.js +171 -0
- package/dist/adapters/integrated-repository-history.d.ts +21 -0
- package/dist/adapters/integrated-repository-history.js +175 -0
- package/dist/adapters/repository-command-diagnostic.d.ts +8 -0
- package/dist/adapters/repository-command-diagnostic.js +46 -0
- package/dist/adapters/repository-command-evidence.d.ts +13 -0
- package/dist/adapters/repository-command-evidence.js +102 -0
- package/dist/adapters/repository-command-runner.d.ts +238 -0
- package/dist/adapters/repository-command-runner.js +1483 -0
- package/dist/adapters/repository-import-context.d.ts +47 -0
- package/dist/adapters/repository-import-context.js +469 -0
- package/dist/adapters/repository-node-test-reporter.d.ts +3 -0
- package/dist/adapters/repository-node-test-reporter.js +27 -0
- package/dist/adapters/repository-review-evidence.d.ts +39 -0
- package/dist/adapters/repository-review-evidence.js +632 -0
- package/dist/adapters/repository-seed-selection.d.ts +7 -0
- package/dist/adapters/repository-seed-selection.js +79 -0
- package/dist/adapters/repository-solution-edits.d.ts +49 -0
- package/dist/adapters/repository-solution-edits.js +136 -0
- package/dist/adapters/repository-vitest-phase-adapter.d.ts +8 -0
- package/dist/adapters/repository-vitest-phase-adapter.js +310 -0
- package/dist/adapters/repository-vitest-reporter.d.ts +24 -0
- package/dist/adapters/repository-vitest-reporter.js +314 -0
- package/dist/adapters/strict-authoring-schema.d.ts +5 -0
- package/dist/adapters/strict-authoring-schema.js +158 -0
- package/dist/adapters/test-discovery.d.ts +30 -0
- package/dist/adapters/test-discovery.js +124 -0
- package/dist/adapters/typescript-repository-index.d.ts +51 -0
- package/dist/adapters/typescript-repository-index.js +226 -0
- package/dist/agentic-capabilities-protocol.d.ts +1373 -0
- package/dist/agentic-capabilities-protocol.js +786 -0
- package/dist/case-checkpoint-store.d.ts +29 -0
- package/dist/case-checkpoint-store.js +133 -0
- package/dist/case-pipeline-protocol-v2.d.ts +184 -0
- package/dist/case-pipeline-protocol-v2.js +193 -0
- package/dist/case-pipeline-protocol.d.ts +2626 -0
- package/dist/case-pipeline-protocol.js +371 -0
- package/dist/effect-api.d.ts +74 -10
- package/dist/effect-api.js +56 -6
- package/dist/errors.d.ts +31 -0
- package/dist/errors.js +10 -0
- package/dist/eval-capability-execution-envelope.d.ts +64 -0
- package/dist/eval-capability-execution-envelope.js +98 -0
- package/dist/eval-capability-policy.d.ts +90 -0
- package/dist/eval-capability-policy.js +107 -0
- package/dist/eval-event-log.d.ts +140 -0
- package/dist/eval-event-log.js +220 -0
- package/dist/evaluation-authoring-policy.d.ts +18 -0
- package/dist/evaluation-authoring-policy.js +19 -0
- package/dist/evaluation-authoring-validation.d.ts +22 -0
- package/dist/evaluation-authoring-validation.js +72 -0
- package/dist/evaluation-evidence.d.ts +20 -0
- package/dist/evaluation-evidence.js +319 -0
- package/dist/evaluation-grader-calibration-protocol.d.ts +108 -0
- package/dist/evaluation-grader-calibration-protocol.js +80 -0
- package/dist/evaluation-grader-calibration.d.ts +18 -0
- package/dist/evaluation-grader-calibration.js +334 -0
- package/dist/evaluation-grading-policy.d.ts +24 -0
- package/dist/evaluation-grading-policy.js +54 -0
- package/dist/evaluation-proposal-policy.d.ts +4 -0
- package/dist/evaluation-proposal-policy.js +91 -0
- package/dist/evaluation-source-retrieval.d.ts +68 -0
- package/dist/evaluation-source-retrieval.js +513 -0
- package/dist/evaluation-structure-policy.d.ts +29 -0
- package/dist/evaluation-structure-policy.js +138 -0
- package/dist/index.d.ts +124 -17
- package/dist/index.js +69 -11
- package/dist/inspection.js +2 -3
- package/dist/project-artifacts.d.ts +7 -2
- package/dist/project-artifacts.js +49 -136
- package/dist/project-authoring.d.ts +66 -5
- package/dist/project-authoring.js +783 -109
- package/dist/project-contracts.d.ts +419 -84
- package/dist/project-contracts.js +160 -52
- package/dist/project-store.js +2 -1
- package/dist/project-workflow.d.ts +5 -4
- package/dist/project-workflow.js +154 -35
- package/dist/repository-adversary-protocol.d.ts +64 -0
- package/dist/repository-adversary-protocol.js +105 -0
- package/dist/repository-behavior-protocol.d.ts +188 -0
- package/dist/repository-behavior-protocol.js +202 -0
- package/dist/repository-benchmark-protocol.d.ts +487 -0
- package/dist/repository-benchmark-protocol.js +96 -0
- package/dist/repository-execution-protocol.d.ts +150 -0
- package/dist/repository-execution-protocol.js +38 -0
- package/dist/repository-fixture-instructions.d.ts +3 -0
- package/dist/repository-fixture-instructions.js +91 -0
- package/dist/repository-fixture-protocol.d.ts +79 -0
- package/dist/repository-fixture-protocol.js +79 -0
- package/dist/repository-foundry-plan-protocol.d.ts +118 -0
- package/dist/repository-foundry-plan-protocol.js +296 -0
- package/dist/repository-foundry-progress-protocol.d.ts +52 -0
- package/dist/repository-foundry-progress-protocol.js +52 -0
- package/dist/repository-improvement-protocol.d.ts +100 -0
- package/dist/repository-improvement-protocol.js +106 -0
- package/dist/repository-language-model-protocol.d.ts +43 -0
- package/dist/repository-language-model-protocol.js +146 -0
- package/dist/repository-oracle-coverage-protocol.d.ts +18 -0
- package/dist/repository-oracle-coverage-protocol.js +39 -0
- package/dist/repository-oracle-execution-binding.d.ts +27 -0
- package/dist/repository-oracle-execution-binding.js +59 -0
- package/dist/repository-oracle-protocol.d.ts +230 -0
- package/dist/repository-oracle-protocol.js +156 -0
- package/dist/repository-oracle-scope-policy.d.ts +22 -0
- package/dist/repository-oracle-scope-policy.js +92 -0
- package/dist/repository-quality-policy.d.ts +15 -0
- package/dist/repository-quality-policy.js +357 -0
- package/dist/repository-routing-benchmark-protocol.d.ts +176 -0
- package/dist/repository-routing-benchmark-protocol.js +103 -0
- package/dist/repository-routing-model-protocol.d.ts +36 -0
- package/dist/repository-routing-model-protocol.js +89 -0
- package/dist/repository-routing-plan-protocol.d.ts +112 -0
- package/dist/repository-routing-plan-protocol.js +58 -0
- package/dist/repository-routing-quality-policy.d.ts +9 -0
- package/dist/repository-routing-quality-policy.js +191 -0
- package/dist/repository-seed-qualification-progress-protocol.d.ts +205 -0
- package/dist/repository-seed-qualification-progress-protocol.js +28 -0
- package/dist/repository-semantic-calibration-protocol.d.ts +768 -0
- package/dist/repository-semantic-calibration-protocol.js +276 -0
- package/dist/repository-semantic-calibration.d.ts +163 -0
- package/dist/repository-semantic-calibration.js +581 -0
- package/dist/repository-specification-contract-facts-protocol.d.ts +224 -0
- package/dist/repository-specification-contract-facts-protocol.js +276 -0
- package/dist/repository-specification-critique-protocol.d.ts +189 -0
- package/dist/repository-specification-critique-protocol.js +103 -0
- package/dist/repository-task-family-protocol.d.ts +24 -0
- package/dist/repository-task-family-protocol.js +37 -0
- package/dist/repository-task-seed-protocol.d.ts +384 -0
- package/dist/repository-task-seed-protocol.js +236 -0
- package/dist/repository-trajectory-protocol.d.ts +20 -0
- package/dist/repository-trajectory-protocol.js +42 -0
- package/dist/service.js +1 -1
- package/dist/services/adversary/service.d.ts +64 -0
- package/dist/services/adversary/service.js +330 -0
- package/dist/services/benchmark-compiler/service.d.ts +450 -0
- package/dist/services/benchmark-compiler/service.js +9 -0
- package/dist/services/budgeted-model/service.d.ts +118 -0
- package/dist/services/budgeted-model/service.js +460 -0
- package/dist/services/case-authoring/service.d.ts +163 -0
- package/dist/services/case-authoring/service.js +1456 -0
- package/dist/services/case-finalization/service.d.ts +283 -0
- package/dist/services/case-finalization/service.js +370 -0
- package/dist/services/case-generation/service.d.ts +619 -0
- package/dist/services/case-generation/service.js +2628 -0
- package/dist/services/case-pipeline/service.d.ts +31 -0
- package/dist/services/case-pipeline/service.js +485 -0
- package/dist/services/case-pipeline-v2/service.d.ts +70 -0
- package/dist/services/case-pipeline-v2/service.js +477 -0
- package/dist/services/command-observability/service.d.ts +13 -0
- package/dist/services/command-observability/service.js +3 -0
- package/dist/services/dimension-labeling/service.d.ts +77 -0
- package/dist/services/dimension-labeling/service.js +188 -0
- package/dist/services/eval-candidate/service.d.ts +208 -0
- package/dist/services/eval-candidate/service.js +64 -0
- package/dist/services/eval-capabilities/service.d.ts +183 -0
- package/dist/services/eval-capabilities/service.js +1433 -0
- package/dist/services/eval-environment/service.d.ts +173 -0
- package/dist/services/eval-environment/service.js +127 -0
- package/dist/services/evidence-reconstruction/service.d.ts +36 -0
- package/dist/services/evidence-reconstruction/service.js +145 -0
- package/dist/services/fixture-builder/service.d.ts +62 -0
- package/dist/services/fixture-builder/service.js +36 -0
- package/dist/services/fixture-validation/service.d.ts +75 -0
- package/dist/services/fixture-validation/service.js +295 -0
- package/dist/services/foundry/service.d.ts +831 -0
- package/dist/services/foundry/service.js +442 -0
- package/dist/services/foundry-progress/service.d.ts +62 -0
- package/dist/services/foundry-progress/service.js +149 -0
- package/dist/services/foundry-v2/service.d.ts +54 -0
- package/dist/services/foundry-v2/service.js +28 -0
- package/dist/services/grounded-authoring/service.d.ts +126 -0
- package/dist/services/grounded-authoring/service.js +822 -0
- package/dist/services/historical-case/service.d.ts +722 -0
- package/dist/services/historical-case/service.js +177 -0
- package/dist/services/improvement-loop/service.d.ts +59 -0
- package/dist/services/improvement-loop/service.js +176 -0
- package/dist/services/language-model/service.d.ts +52 -0
- package/dist/services/language-model/service.js +194 -0
- package/dist/services/oracle-builder/service.d.ts +146 -0
- package/dist/services/oracle-builder/service.js +513 -0
- package/dist/services/oracle-coverage/service.d.ts +28 -0
- package/dist/services/oracle-coverage/service.js +50 -0
- package/dist/services/oracle-coverage-witness/service.d.ts +130 -0
- package/dist/services/oracle-coverage-witness/service.js +538 -0
- package/dist/services/pipeline-challenge/service.d.ts +551 -0
- package/dist/services/pipeline-challenge/service.js +427 -0
- package/dist/services/pipeline-controls/service.d.ts +130 -0
- package/dist/services/pipeline-controls/service.js +483 -0
- package/dist/services/pipeline-oracle/service.d.ts +8 -0
- package/dist/services/pipeline-oracle/service.js +256 -0
- package/dist/services/pipeline-seed/service.d.ts +298 -0
- package/dist/services/pipeline-seed/service.js +428 -0
- package/dist/services/pipeline-spec/service.d.ts +103 -0
- package/dist/services/pipeline-spec/service.js +619 -0
- package/dist/services/pipeline-tournament/service.d.ts +258 -0
- package/dist/services/pipeline-tournament/service.js +476 -0
- package/dist/services/quality-gate/service.d.ts +233 -0
- package/dist/services/quality-gate/service.js +136 -0
- package/dist/services/repository-bundle/service.d.ts +33 -0
- package/dist/services/repository-bundle/service.js +114 -0
- package/dist/services/repository-model/service.d.ts +105 -0
- package/dist/services/repository-model/service.js +250 -0
- package/dist/services/repository-public-artifact/service.d.ts +133 -0
- package/dist/services/repository-public-artifact/service.js +330 -0
- package/dist/services/routing-benchmark/service.d.ts +362 -0
- package/dist/services/routing-benchmark/service.js +96 -0
- package/dist/services/specification-critic/service.d.ts +92 -0
- package/dist/services/specification-critic/service.js +172 -0
- package/dist/services/task-family/service.d.ts +40 -0
- package/dist/services/task-family/service.js +55 -0
- package/dist/services/task-seed/service.d.ts +906 -0
- package/dist/services/task-seed/service.js +1406 -0
- package/dist/services/task-specification/service.d.ts +27 -0
- package/dist/services/task-specification/service.js +40 -0
- package/dist/services/trajectory-policy/service.d.ts +110 -0
- package/dist/services/trajectory-policy/service.js +216 -0
- package/dist/test/agentic-capabilities-protocol.test.d.ts +1 -0
- package/dist/test/agentic-capabilities-protocol.test.js +570 -0
- package/dist/test/agentic-capabilities.test.d.ts +1 -0
- package/dist/test/agentic-capabilities.test.js +1461 -0
- package/dist/test/agentic-environment.test.d.ts +1 -0
- package/dist/test/agentic-environment.test.js +213 -0
- package/dist/test/case-pipeline-foundation.test.d.ts +1 -0
- package/dist/test/case-pipeline-foundation.test.js +535 -0
- package/dist/test/case-pipeline-protocol-v2.test.d.ts +1 -0
- package/dist/test/case-pipeline-protocol-v2.test.js +124 -0
- package/dist/test/case-pipeline-v2.test.d.ts +1 -0
- package/dist/test/case-pipeline-v2.test.js +286 -0
- package/dist/test/case-pipeline.test.d.ts +1 -0
- package/dist/test/case-pipeline.test.js +851 -0
- package/dist/test/eval-capability-policy.test.d.ts +1 -0
- package/dist/test/eval-capability-policy.test.js +50 -0
- package/dist/test/eval-event-log.test.d.ts +1 -0
- package/dist/test/eval-event-log.test.js +125 -0
- package/dist/test/evaluation-evidence-freshness.test.d.ts +1 -0
- package/dist/test/evaluation-evidence-freshness.test.js +44 -0
- package/dist/test/evaluation-evidence.test.d.ts +1 -0
- package/dist/test/evaluation-evidence.test.js +230 -0
- package/dist/test/evaluation-grader-calibration.test.d.ts +1 -0
- package/dist/test/evaluation-grader-calibration.test.js +373 -0
- package/dist/test/evaluation-proposal-digest.test.d.ts +1 -0
- package/dist/test/evaluation-proposal-digest.test.js +187 -0
- package/dist/test/evaluation-source-retrieval.test.d.ts +1 -0
- package/dist/test/evaluation-source-retrieval.test.js +237 -0
- package/dist/test/evaluation-structure-policy.test.d.ts +1 -0
- package/dist/test/evaluation-structure-policy.test.js +196 -0
- package/dist/test/fixtures/repository-resource-panel.d.ts +39 -0
- package/dist/test/fixtures/repository-resource-panel.js +111 -0
- package/dist/test/fixtures/vitest-boundary-panel.d.ts +84 -0
- package/dist/test/fixtures/vitest-boundary-panel.js +120 -0
- package/dist/test/fixtures/vitest-phase-panel.d.ts +135 -0
- package/dist/test/fixtures/vitest-phase-panel.js +213 -0
- package/dist/test/fixtures/vitest-reporter-results.d.ts +76 -0
- package/dist/test/fixtures/vitest-reporter-results.js +94 -0
- package/dist/test/grounded-authoring.test.d.ts +1 -0
- package/dist/test/grounded-authoring.test.js +565 -0
- package/dist/test/integrated-repository-history.test.d.ts +1 -0
- package/dist/test/integrated-repository-history.test.js +227 -0
- package/dist/test/project-authoring.test.js +593 -43
- package/dist/test/project-workflow.test.js +419 -40
- package/dist/test/repository-authoring-artifacts.test.d.ts +1 -0
- package/dist/test/repository-authoring-artifacts.test.js +185 -0
- package/dist/test/repository-bundle.test.d.ts +1 -0
- package/dist/test/repository-bundle.test.js +52 -0
- package/dist/test/repository-case-generation.test.d.ts +1 -0
- package/dist/test/repository-case-generation.test.js +3465 -0
- package/dist/test/repository-command-diagnostic.test.d.ts +1 -0
- package/dist/test/repository-command-diagnostic.test.js +55 -0
- package/dist/test/repository-command-signals.test.d.ts +1 -0
- package/dist/test/repository-command-signals.test.js +124 -0
- package/dist/test/repository-fixture-scope-coverage.test.d.ts +1 -0
- package/dist/test/repository-fixture-scope-coverage.test.js +127 -0
- package/dist/test/repository-fixture-validation.test.d.ts +1 -0
- package/dist/test/repository-fixture-validation.test.js +362 -0
- package/dist/test/repository-foundry-progress.test.d.ts +1 -0
- package/dist/test/repository-foundry-progress.test.js +110 -0
- package/dist/test/repository-foundry-quality.test.d.ts +1 -0
- package/dist/test/repository-foundry-quality.test.js +1138 -0
- package/dist/test/repository-import-context.test.d.ts +1 -0
- package/dist/test/repository-import-context.test.js +354 -0
- package/dist/test/repository-model-authoring.test.d.ts +1 -0
- package/dist/test/repository-model-authoring.test.js +544 -0
- package/dist/test/repository-model.test.d.ts +1 -0
- package/dist/test/repository-model.test.js +2195 -0
- package/dist/test/repository-node-test-reporter.test.d.ts +1 -0
- package/dist/test/repository-node-test-reporter.test.js +104 -0
- package/dist/test/repository-oracle-concurrency.test.d.ts +1 -0
- package/dist/test/repository-oracle-concurrency.test.js +542 -0
- package/dist/test/repository-oracle-coverage-witness.test.d.ts +1 -0
- package/dist/test/repository-oracle-coverage-witness.test.js +511 -0
- package/dist/test/repository-oracle-coverage.test.d.ts +1 -0
- package/dist/test/repository-oracle-coverage.test.js +168 -0
- package/dist/test/repository-oracle-evidence.test.d.ts +1 -0
- package/dist/test/repository-oracle-evidence.test.js +185 -0
- package/dist/test/repository-oracle-plan.test.d.ts +1 -0
- package/dist/test/repository-oracle-plan.test.js +176 -0
- package/dist/test/repository-overlay-isolation.test.d.ts +1 -0
- package/dist/test/repository-overlay-isolation.test.js +85 -0
- package/dist/test/repository-preparation-cache.test.d.ts +1 -0
- package/dist/test/repository-preparation-cache.test.js +414 -0
- package/dist/test/repository-public-artifact.test.d.ts +1 -0
- package/dist/test/repository-public-artifact.test.js +273 -0
- package/dist/test/repository-qualification-diagnostics.test.d.ts +1 -0
- package/dist/test/repository-qualification-diagnostics.test.js +524 -0
- package/dist/test/repository-reference-authoring.test.d.ts +1 -0
- package/dist/test/repository-reference-authoring.test.js +1633 -0
- package/dist/test/repository-review-evidence-v2.test.d.ts +1 -0
- package/dist/test/repository-review-evidence-v2.test.js +183 -0
- package/dist/test/repository-review-evidence.test.d.ts +1 -0
- package/dist/test/repository-review-evidence.test.js +124 -0
- package/dist/test/repository-seed-exclusions.test.d.ts +1 -0
- package/dist/test/repository-seed-exclusions.test.js +96 -0
- package/dist/test/repository-seed-selection.test.d.ts +1 -0
- package/dist/test/repository-seed-selection.test.js +504 -0
- package/dist/test/repository-semantic-calibration.test.d.ts +1 -0
- package/dist/test/repository-semantic-calibration.test.js +688 -0
- package/dist/test/repository-solution-edits.test.d.ts +1 -0
- package/dist/test/repository-solution-edits.test.js +377 -0
- package/dist/test/repository-specification-budget.test.d.ts +1 -0
- package/dist/test/repository-specification-budget.test.js +171 -0
- package/dist/test/repository-specification-contract-checkpoint.test.d.ts +1 -0
- package/dist/test/repository-specification-contract-checkpoint.test.js +228 -0
- package/dist/test/repository-specification-contract-facts.test.d.ts +1 -0
- package/dist/test/repository-specification-contract-facts.test.js +177 -0
- package/dist/test/repository-trajectory-authoring.test.d.ts +1 -0
- package/dist/test/repository-trajectory-authoring.test.js +176 -0
- package/dist/test/repository-valid-control-plan.test.d.ts +1 -0
- package/dist/test/repository-valid-control-plan.test.js +45 -0
- package/dist/test/repository-vitest-phase.test.d.ts +1 -0
- package/dist/test/repository-vitest-phase.test.js +848 -0
- package/dist/test/repository-vitest-reporter.test.d.ts +1 -0
- package/dist/test/repository-vitest-reporter.test.js +158 -0
- package/dist/test/repository-workspace-build.test.d.ts +1 -0
- package/dist/test/repository-workspace-build.test.js +160 -0
- package/dist/test/strict-authoring-schema.test.d.ts +1 -0
- package/dist/test/strict-authoring-schema.test.js +169 -0
- package/package.json +48 -6
|
@@ -0,0 +1,371 @@
|
|
|
1
|
+
import { createHash } from "node:crypto";
|
|
2
|
+
import { Data, Schema } from "effect";
|
|
3
|
+
import { RepositoryBenchmarkCaseV1, RepositoryCandidateVisibleTaskV1 } from "./repository-benchmark-protocol.js";
|
|
4
|
+
import { RepositoryHiddenFixtureSuiteV1, RepositoryOracleFixtureV1 } from "./repository-fixture-protocol.js";
|
|
5
|
+
import { RepositoryFoundryModelPlanV1, RepositoryFoundryReasoningEffortV1, repositoryFoundryModelPlanV1 } from "./repository-language-model-protocol.js";
|
|
6
|
+
import { RepositoryHiddenOracleV1, RepositoryOracleAdequacyReportV1, RepositoryReferenceSolutionV1 } from "./repository-oracle-protocol.js";
|
|
7
|
+
import { RepositorySeedCrossBatchQualificationV1, RepositoryTaskSeedV1 } from "./repository-task-seed-protocol.js";
|
|
8
|
+
/**
|
|
9
|
+
* Additive case-generation pipeline protocol. Every stage persists one
|
|
10
|
+
* checkpoint; the orchestrator resumes from the first stage without one.
|
|
11
|
+
* Model output is advisory everywhere; executed observations are the only
|
|
12
|
+
* admission evidence.
|
|
13
|
+
*/
|
|
14
|
+
export const PIPELINE_STAGES = [
|
|
15
|
+
"seed",
|
|
16
|
+
"spec",
|
|
17
|
+
"oracle",
|
|
18
|
+
"controls",
|
|
19
|
+
"tournament",
|
|
20
|
+
"challenge",
|
|
21
|
+
"admit"
|
|
22
|
+
];
|
|
23
|
+
export const PipelineStageV1 = Schema.Literals(PIPELINE_STAGES);
|
|
24
|
+
export const PipelineStatusV1 = Schema.Literals([
|
|
25
|
+
"running",
|
|
26
|
+
"admitted",
|
|
27
|
+
"rejected",
|
|
28
|
+
"exhausted",
|
|
29
|
+
"failed"
|
|
30
|
+
]);
|
|
31
|
+
export const DimensionContractV1 = Schema.Struct({
|
|
32
|
+
id: Schema.String,
|
|
33
|
+
description: Schema.String,
|
|
34
|
+
includes: Schema.Array(Schema.String),
|
|
35
|
+
excludes: Schema.Array(Schema.String)
|
|
36
|
+
});
|
|
37
|
+
/** The single budget. No per-stage or per-role allowances exist. */
|
|
38
|
+
export const PipelineBudgetV1 = Schema.Struct({
|
|
39
|
+
maxModelCalls: Schema.Finite,
|
|
40
|
+
maxWallMs: Schema.Finite
|
|
41
|
+
});
|
|
42
|
+
export const PipelineStateV1 = Schema.Struct({
|
|
43
|
+
version: Schema.Literal(1),
|
|
44
|
+
caseId: Schema.String,
|
|
45
|
+
repositoryRoot: Schema.String,
|
|
46
|
+
dimension: DimensionContractV1,
|
|
47
|
+
model: Schema.String,
|
|
48
|
+
/**
|
|
49
|
+
* Digest of the immutable scientific inputs. The requested cumulative budget
|
|
50
|
+
* is deliberately excluded so an exhausted case can resume with a larger
|
|
51
|
+
* ceiling without changing its identity.
|
|
52
|
+
*/
|
|
53
|
+
identityDigest: Schema.optionalKey(Schema.String),
|
|
54
|
+
stage: Schema.Literals([...PIPELINE_STAGES, "done"]),
|
|
55
|
+
status: PipelineStatusV1,
|
|
56
|
+
reason: Schema.optionalKey(Schema.String),
|
|
57
|
+
startedAt: Schema.String,
|
|
58
|
+
updatedAt: Schema.String,
|
|
59
|
+
modelCalls: Schema.Finite,
|
|
60
|
+
/** Cumulative across resumes; the budget wrapper persists it on every received call. */
|
|
61
|
+
elapsedMs: Schema.Finite,
|
|
62
|
+
resumeCount: Schema.Finite,
|
|
63
|
+
folds: Schema.Finite
|
|
64
|
+
});
|
|
65
|
+
export const PipelineLedgerPhaseV1 = Schema.Literals([
|
|
66
|
+
"start",
|
|
67
|
+
"received",
|
|
68
|
+
"validated",
|
|
69
|
+
"failed",
|
|
70
|
+
"retry"
|
|
71
|
+
]);
|
|
72
|
+
export const PipelineLedgerEntryV1 = Schema.Struct({
|
|
73
|
+
at: Schema.String,
|
|
74
|
+
operationId: Schema.String,
|
|
75
|
+
role: Schema.String,
|
|
76
|
+
model: Schema.String,
|
|
77
|
+
phase: PipelineLedgerPhaseV1,
|
|
78
|
+
attempt: Schema.Finite,
|
|
79
|
+
callId: Schema.optionalKey(Schema.String),
|
|
80
|
+
detail: Schema.optionalKey(Schema.String),
|
|
81
|
+
inputBytes: Schema.optionalKey(Schema.Finite),
|
|
82
|
+
outputBytes: Schema.optionalKey(Schema.Finite)
|
|
83
|
+
});
|
|
84
|
+
export const PIPELINE_DEFAULT_ORACLE_REPETITIONS = 2;
|
|
85
|
+
export const PIPELINE_DEFAULT_ORACLE_CONCURRENCY = 2;
|
|
86
|
+
export const PIPELINE_DEFAULT_FOLD_LIMIT = 2;
|
|
87
|
+
export const PipelineCaseContextV1 = Schema.Struct({
|
|
88
|
+
repositoryRoot: Schema.String,
|
|
89
|
+
caseId: Schema.String,
|
|
90
|
+
caseDirectory: Schema.String,
|
|
91
|
+
dimension: DimensionContractV1,
|
|
92
|
+
primaryModel: Schema.String,
|
|
93
|
+
reasoningEffort: RepositoryFoundryReasoningEffortV1,
|
|
94
|
+
/** Legacy plan so substrate services that still take a plan keep working. */
|
|
95
|
+
modelPlan: RepositoryFoundryModelPlanV1,
|
|
96
|
+
budget: PipelineBudgetV1,
|
|
97
|
+
referenceCommit: Schema.optionalKey(Schema.String),
|
|
98
|
+
requestedRef: Schema.optionalKey(Schema.String),
|
|
99
|
+
oracleRepetitions: Schema.Finite,
|
|
100
|
+
oracleConcurrency: Schema.Finite,
|
|
101
|
+
foldLimit: Schema.Finite
|
|
102
|
+
});
|
|
103
|
+
/**
|
|
104
|
+
* Build a context with defaults and a legacy model plan. Legacy primary roles
|
|
105
|
+
* require high or stronger reasoning, so weaker efforts are lifted to "high"
|
|
106
|
+
* inside the plan only; `reasoningEffort` keeps the requested value.
|
|
107
|
+
*/
|
|
108
|
+
export const makePipelineCaseContextV1 = (input) => {
|
|
109
|
+
const reasoningEffort = input.reasoningEffort ?? "high";
|
|
110
|
+
const planEffort = reasoningEffort === "low" || reasoningEffort === "medium" ? "high" : reasoningEffort;
|
|
111
|
+
return {
|
|
112
|
+
repositoryRoot: input.repositoryRoot,
|
|
113
|
+
caseId: input.caseId,
|
|
114
|
+
caseDirectory: input.caseDirectory,
|
|
115
|
+
dimension: input.dimension,
|
|
116
|
+
primaryModel: input.primaryModel,
|
|
117
|
+
reasoningEffort,
|
|
118
|
+
modelPlan: repositoryFoundryModelPlanV1({
|
|
119
|
+
primaryModel: input.primaryModel,
|
|
120
|
+
oracleCriticModel: input.primaryModel,
|
|
121
|
+
primaryReasoningEffort: planEffort,
|
|
122
|
+
reviewReasoningEffort: planEffort
|
|
123
|
+
}),
|
|
124
|
+
budget: input.budget,
|
|
125
|
+
...(input.referenceCommit === undefined ? {} : { referenceCommit: input.referenceCommit }),
|
|
126
|
+
...(input.requestedRef === undefined ? {} : { requestedRef: input.requestedRef }),
|
|
127
|
+
oracleRepetitions: input.oracleRepetitions ?? PIPELINE_DEFAULT_ORACLE_REPETITIONS,
|
|
128
|
+
oracleConcurrency: input.oracleConcurrency ?? PIPELINE_DEFAULT_ORACLE_CONCURRENCY,
|
|
129
|
+
foldLimit: input.foldLimit ?? PIPELINE_DEFAULT_FOLD_LIMIT
|
|
130
|
+
};
|
|
131
|
+
};
|
|
132
|
+
// Schemas for substrate values that only exist as TypeScript types today.
|
|
133
|
+
export const RepositoryCommandFileOverrideSchemaV1 = Schema.Union([
|
|
134
|
+
Schema.Struct({ path: Schema.String, content: Schema.String }),
|
|
135
|
+
Schema.Struct({ path: Schema.String, delete: Schema.Literal(true) })
|
|
136
|
+
]);
|
|
137
|
+
export const RepositoryExecutableSolutionSchemaV1 = Schema.Struct({
|
|
138
|
+
...RepositoryReferenceSolutionV1.fields,
|
|
139
|
+
commit: Schema.String,
|
|
140
|
+
fileOverrides: Schema.optionalKey(Schema.Array(RepositoryCommandFileOverrideSchemaV1))
|
|
141
|
+
});
|
|
142
|
+
const RepositoryTestEvidenceSchemaV1 = Schema.Struct({
|
|
143
|
+
outcome: Schema.Literals(["pass", "fail", "unknown"]),
|
|
144
|
+
detail: Schema.String,
|
|
145
|
+
inconclusiveReason: Schema.optionalKey(Schema.Literal("unattributed-test-failure"))
|
|
146
|
+
});
|
|
147
|
+
const RepositoryFixtureOverlaySchemaV1 = Schema.Struct({
|
|
148
|
+
path: Schema.String,
|
|
149
|
+
content: Schema.String,
|
|
150
|
+
fixtureIds: Schema.Array(Schema.String)
|
|
151
|
+
});
|
|
152
|
+
export const RepositoryOracleCoverageWitnessSchemaV1 = Schema.Struct({
|
|
153
|
+
scopeId: Schema.String,
|
|
154
|
+
hypothesis: Schema.String,
|
|
155
|
+
outcome: Schema.Literals(["already-detected", "confirmed-escape"]),
|
|
156
|
+
explanation: Schema.String,
|
|
157
|
+
referenceCommit: Schema.String,
|
|
158
|
+
fileOverrides: Schema.Array(Schema.Struct({ path: Schema.String, content: Schema.String })),
|
|
159
|
+
fixture: RepositoryOracleFixtureV1,
|
|
160
|
+
overlay: RepositoryFixtureOverlaySchemaV1,
|
|
161
|
+
referenceEvidence: RepositoryTestEvidenceSchemaV1,
|
|
162
|
+
mutantEvidence: RepositoryTestEvidenceSchemaV1,
|
|
163
|
+
incumbentEvidence: Schema.Array(Schema.Struct({ fixtureId: Schema.String, evidence: RepositoryTestEvidenceSchemaV1 })),
|
|
164
|
+
executionOperationId: Schema.String
|
|
165
|
+
});
|
|
166
|
+
const checkpointHeader = (stage) => ({
|
|
167
|
+
version: Schema.Literal(1),
|
|
168
|
+
stage: Schema.Literal(stage),
|
|
169
|
+
/** sha256 of the previous checkpoint's stable JSON, or of the context for seed. */
|
|
170
|
+
parentDigest: Schema.String,
|
|
171
|
+
createdAt: Schema.String
|
|
172
|
+
});
|
|
173
|
+
export const SeedCheckpointV1 = Schema.Struct({
|
|
174
|
+
...checkpointHeader("seed"),
|
|
175
|
+
seed: RepositoryTaskSeedV1,
|
|
176
|
+
qualification: RepositorySeedCrossBatchQualificationV1,
|
|
177
|
+
repository: Schema.Struct({
|
|
178
|
+
root: Schema.String,
|
|
179
|
+
commit: Schema.String,
|
|
180
|
+
tree: Schema.String
|
|
181
|
+
}),
|
|
182
|
+
selection: Schema.Struct({
|
|
183
|
+
candidates: Schema.Array(Schema.Struct({
|
|
184
|
+
commit: Schema.String,
|
|
185
|
+
subject: Schema.String,
|
|
186
|
+
score: Schema.Finite,
|
|
187
|
+
rationale: Schema.String
|
|
188
|
+
})),
|
|
189
|
+
selectedCommit: Schema.String,
|
|
190
|
+
rejected: Schema.Array(Schema.Struct({ commit: Schema.String, reason: Schema.String }))
|
|
191
|
+
})
|
|
192
|
+
});
|
|
193
|
+
export const SpecScopeEntryV1 = Schema.Struct({
|
|
194
|
+
id: Schema.String,
|
|
195
|
+
kind: Schema.Literals(["changed", "preserved"]),
|
|
196
|
+
behaviorIds: Schema.Array(Schema.String),
|
|
197
|
+
description: Schema.String
|
|
198
|
+
});
|
|
199
|
+
export const SpecCheckpointV1 = Schema.Struct({
|
|
200
|
+
...checkpointHeader("spec"),
|
|
201
|
+
visible: RepositoryCandidateVisibleTaskV1,
|
|
202
|
+
analysis: Schema.Struct({
|
|
203
|
+
summary: Schema.String,
|
|
204
|
+
userObservableBehaviors: Schema.Array(Schema.String),
|
|
205
|
+
constraints: Schema.Array(Schema.String),
|
|
206
|
+
risks: Schema.Array(Schema.String)
|
|
207
|
+
}),
|
|
208
|
+
scope: Schema.Array(SpecScopeEntryV1),
|
|
209
|
+
contextSources: Schema.Array(Schema.Struct({ path: Schema.String, content: Schema.String })),
|
|
210
|
+
/** Advisory model reviews; write RepositorySpecificationReviewV1 values where possible. */
|
|
211
|
+
reviews: Schema.Array(Schema.Unknown),
|
|
212
|
+
revisions: Schema.Finite,
|
|
213
|
+
leakFindings: Schema.Array(Schema.String)
|
|
214
|
+
});
|
|
215
|
+
export const OracleFixtureMetadataV1 = Schema.Struct({
|
|
216
|
+
id: Schema.String,
|
|
217
|
+
kind: Schema.Literals(["historical-regression", "boundary", "counterfactual", "metamorphic"]),
|
|
218
|
+
expectationMode: Schema.Literals(["changes", "preserved"]),
|
|
219
|
+
description: Schema.String,
|
|
220
|
+
expectedBehavior: Schema.String,
|
|
221
|
+
testPath: Schema.String
|
|
222
|
+
});
|
|
223
|
+
export const OracleScopeCoverageV1 = Schema.Struct({
|
|
224
|
+
scopeId: Schema.String,
|
|
225
|
+
outcome: Schema.Literals(["covered", "unsupported", "contradictory", "uncertain"]),
|
|
226
|
+
fixtureIds: Schema.Array(Schema.String),
|
|
227
|
+
detail: Schema.String
|
|
228
|
+
});
|
|
229
|
+
export const OracleCheckpointV1 = Schema.Struct({
|
|
230
|
+
...checkpointHeader("oracle"),
|
|
231
|
+
suite: RepositoryHiddenFixtureSuiteV1,
|
|
232
|
+
fixtureMetadata: Schema.Array(OracleFixtureMetadataV1),
|
|
233
|
+
/** Advisory model self-report; witnesses are the executed evidence. */
|
|
234
|
+
scopeCoverage: Schema.Array(OracleScopeCoverageV1),
|
|
235
|
+
preflight: Schema.Struct({ passed: Schema.Literal(true), repairs: Schema.Finite }),
|
|
236
|
+
witnesses: Schema.Array(RepositoryOracleCoverageWitnessSchemaV1),
|
|
237
|
+
coverageRounds: Schema.Finite
|
|
238
|
+
});
|
|
239
|
+
export const SolutionCritiqueV1 = Schema.Struct({
|
|
240
|
+
solutionId: Schema.String,
|
|
241
|
+
classification: Schema.Literals(["valid", "wrong", "uncertain"]),
|
|
242
|
+
detail: Schema.String,
|
|
243
|
+
violatedBehaviorIds: Schema.Array(Schema.String)
|
|
244
|
+
});
|
|
245
|
+
export const ControlsCheckpointV1 = Schema.Struct({
|
|
246
|
+
...checkpointHeader("controls"),
|
|
247
|
+
valid: Schema.Array(RepositoryExecutableSolutionSchemaV1),
|
|
248
|
+
wrong: Schema.Array(RepositoryExecutableSolutionSchemaV1),
|
|
249
|
+
/** Advisory. */
|
|
250
|
+
critiques: Schema.Array(SolutionCritiqueV1),
|
|
251
|
+
regenerations: Schema.Finite
|
|
252
|
+
});
|
|
253
|
+
export const TournamentHistoryEntryV1 = Schema.Struct({
|
|
254
|
+
sequence: Schema.Finite,
|
|
255
|
+
decision: Schema.Literals(["accepted", "rejected"]),
|
|
256
|
+
targeted: Schema.Array(Schema.String),
|
|
257
|
+
rejectionReasons: Schema.Array(Schema.String),
|
|
258
|
+
candidateAdequacy: Schema.optionalKey(RepositoryOracleAdequacyReportV1)
|
|
259
|
+
});
|
|
260
|
+
export const TournamentCheckpointV1 = Schema.Struct({
|
|
261
|
+
...checkpointHeader("tournament"),
|
|
262
|
+
/** The incumbent suite after the last accepted iteration. */
|
|
263
|
+
suite: RepositoryHiddenFixtureSuiteV1,
|
|
264
|
+
adequacy: RepositoryOracleAdequacyReportV1,
|
|
265
|
+
oracle: RepositoryHiddenOracleV1,
|
|
266
|
+
valid: Schema.Array(RepositoryExecutableSolutionSchemaV1),
|
|
267
|
+
wrong: Schema.Array(RepositoryExecutableSolutionSchemaV1),
|
|
268
|
+
/** Lazy advisory control reviews, checkpointed after every completed critic round. */
|
|
269
|
+
critiques: Schema.optionalKey(Schema.Array(SolutionCritiqueV1)),
|
|
270
|
+
history: Schema.Array(TournamentHistoryEntryV1),
|
|
271
|
+
iterations: Schema.Finite
|
|
272
|
+
});
|
|
273
|
+
export const ChallengeCheckpointV1 = Schema.Struct({
|
|
274
|
+
...checkpointHeader("challenge"),
|
|
275
|
+
heldOut: Schema.Array(RepositoryExecutableSolutionSchemaV1),
|
|
276
|
+
adequacy: RepositoryOracleAdequacyReportV1,
|
|
277
|
+
oracle: RepositoryHiddenOracleV1,
|
|
278
|
+
verdict: Schema.Literals(["passed", "failed"]),
|
|
279
|
+
fold: Schema.Finite
|
|
280
|
+
});
|
|
281
|
+
export const AdmitCheckpointV1 = Schema.Struct({
|
|
282
|
+
...checkpointHeader("admit"),
|
|
283
|
+
benchmarkCase: RepositoryBenchmarkCaseV1,
|
|
284
|
+
verdict: Schema.Literals(["admitted", "rejected"]),
|
|
285
|
+
reasons: Schema.Array(Schema.String),
|
|
286
|
+
evidence: Schema.Struct({
|
|
287
|
+
validCount: Schema.Finite,
|
|
288
|
+
wrongCount: Schema.Finite,
|
|
289
|
+
heldOutWrongCount: Schema.Finite,
|
|
290
|
+
semanticKills: Schema.Array(Schema.String)
|
|
291
|
+
})
|
|
292
|
+
});
|
|
293
|
+
export const PIPELINE_CHECKPOINT_SCHEMAS = {
|
|
294
|
+
seed: SeedCheckpointV1,
|
|
295
|
+
spec: SpecCheckpointV1,
|
|
296
|
+
oracle: OracleCheckpointV1,
|
|
297
|
+
controls: ControlsCheckpointV1,
|
|
298
|
+
tournament: TournamentCheckpointV1,
|
|
299
|
+
challenge: ChallengeCheckpointV1,
|
|
300
|
+
admit: AdmitCheckpointV1
|
|
301
|
+
};
|
|
302
|
+
export const HostFeedbackFindingKindV1 = Schema.Literals([
|
|
303
|
+
"schema",
|
|
304
|
+
"contract",
|
|
305
|
+
"materialization",
|
|
306
|
+
"compile",
|
|
307
|
+
"execution",
|
|
308
|
+
"critique"
|
|
309
|
+
]);
|
|
310
|
+
export const HostFeedbackFindingV1 = Schema.Struct({
|
|
311
|
+
kind: HostFeedbackFindingKindV1,
|
|
312
|
+
message: Schema.String,
|
|
313
|
+
path: Schema.optionalKey(Schema.String)
|
|
314
|
+
});
|
|
315
|
+
/** Sent back to the SAME role with its prior output; never a terminal outcome. */
|
|
316
|
+
export const HostFeedbackV1 = Schema.Struct({
|
|
317
|
+
attempt: Schema.Finite,
|
|
318
|
+
findings: Schema.Array(HostFeedbackFindingV1),
|
|
319
|
+
priorOutput: Schema.Unknown
|
|
320
|
+
});
|
|
321
|
+
export class PipelineBudgetExhaustedError extends Data.TaggedError("PipelineBudgetExhaustedError") {
|
|
322
|
+
get message() {
|
|
323
|
+
return (`RouteKit case pipeline exhausted its budget during ${this.stage}: ` +
|
|
324
|
+
`modelCalls=${String(this.modelCalls)}/${String(this.budget.maxModelCalls)} ` +
|
|
325
|
+
`elapsedMs=${String(this.elapsedMs)}/${String(this.budget.maxWallMs)}`);
|
|
326
|
+
}
|
|
327
|
+
}
|
|
328
|
+
export class PipelineStageError extends Data.TaggedError("PipelineStageError") {
|
|
329
|
+
get message() {
|
|
330
|
+
return `RouteKit case pipeline stage ${this.stage} ${this.terminal ? "failed" : "did not complete"}: ${this.detail}`;
|
|
331
|
+
}
|
|
332
|
+
}
|
|
333
|
+
const stableJson = (value) => JSON.stringify(value, (_key, entry) => {
|
|
334
|
+
if (entry === null || typeof entry !== "object" || Array.isArray(entry))
|
|
335
|
+
return entry;
|
|
336
|
+
const record = entry;
|
|
337
|
+
return Object.fromEntries(Object.keys(record)
|
|
338
|
+
.sort()
|
|
339
|
+
.map((key) => [key, record[key]]));
|
|
340
|
+
});
|
|
341
|
+
/** sha256 over key-sorted JSON so equal checkpoints digest equally. */
|
|
342
|
+
export const checkpointDigestV1 = (value) => createHash("sha256").update(stableJson(value)).digest("hex");
|
|
343
|
+
/** Stable admission identity; cumulative budget is intentionally extendable. */
|
|
344
|
+
export const pipelineScientificIdentityDigestV1 = (context, map) => checkpointDigestV1({
|
|
345
|
+
version: 1,
|
|
346
|
+
caseId: context.caseId,
|
|
347
|
+
repositoryRoot: context.repositoryRoot,
|
|
348
|
+
caseDirectory: context.caseDirectory,
|
|
349
|
+
dimension: context.dimension,
|
|
350
|
+
primaryModel: context.primaryModel,
|
|
351
|
+
reasoningEffort: context.reasoningEffort,
|
|
352
|
+
modelPlan: context.modelPlan,
|
|
353
|
+
referenceCommit: context.referenceCommit,
|
|
354
|
+
requestedRef: context.requestedRef,
|
|
355
|
+
oracleRepetitions: context.oracleRepetitions,
|
|
356
|
+
oracleConcurrency: context.oracleConcurrency,
|
|
357
|
+
foldLimit: context.foldLimit,
|
|
358
|
+
repository: map.repository
|
|
359
|
+
});
|
|
360
|
+
const PROGRESS_STAGE_BY_PIPELINE_STAGE = {
|
|
361
|
+
seed: "replay-qualify",
|
|
362
|
+
spec: "author-specification",
|
|
363
|
+
oracle: "oracle-design",
|
|
364
|
+
controls: "valid-solutions",
|
|
365
|
+
tournament: "tournament",
|
|
366
|
+
challenge: "held-out-tournament",
|
|
367
|
+
admit: "admit"
|
|
368
|
+
};
|
|
369
|
+
export const pipelineStageToProgressStageV1 = (stage) => PROGRESS_STAGE_BY_PIPELINE_STAGE[stage];
|
|
370
|
+
export const pipelineStageIndexV1 = (stage) => PIPELINE_STAGES.indexOf(stage);
|
|
371
|
+
export const nextPipelineStageV1 = (stage) => PIPELINE_STAGES[pipelineStageIndexV1(stage) + 1] ?? "done";
|
package/dist/effect-api.d.ts
CHANGED
|
@@ -1,22 +1,86 @@
|
|
|
1
|
-
export {
|
|
1
|
+
export { evalAuthoringRequestByteLimit, evalAuthoringResponsesRequestBody } from "./adapters/authoring-responses-request.js";
|
|
2
2
|
export { captureFrozenRepositorySnapshotV3 } from "./adapters/git-basis-snapshot.js";
|
|
3
|
-
export
|
|
4
|
-
export {
|
|
5
|
-
export type
|
|
6
|
-
export {
|
|
7
|
-
export
|
|
8
|
-
export {
|
|
3
|
+
export { captureGitTreeSnapshotV1, readGitChangeEpisodesV1, readGitTreeFileV1 } from "./adapters/git-task-history.js";
|
|
4
|
+
export { inspectRepositorySolutionPathsV1, replayRepositoryCommandsV1, replayRepositoryCommandV1 } from "./adapters/repository-command-runner.js";
|
|
5
|
+
export { decodeRepositoryReviewEvidenceV1, decodeRepositoryReviewEvidenceV2, encodeRepositoryReviewEvidenceV1, encodeRepositoryReviewEvidenceV2, REPOSITORY_REVIEW_EVIDENCE_INSTRUCTIONS, REPOSITORY_REVIEW_EVIDENCE_V2_INSTRUCTIONS, type RepositoryReviewEvidenceV1, type RepositoryReviewEvidenceV2 } from "./adapters/repository-review-evidence.js";
|
|
6
|
+
export { discoverRepositoryTestsV1 } from "./adapters/test-discovery.js";
|
|
7
|
+
export { indexTypeScriptRepositoryV1 } from "./adapters/typescript-repository-index.js";
|
|
8
|
+
export { projectRoutingBasisV3, projectRoutingDimensionV3 } from "./basis-projection.js";
|
|
9
|
+
export type { PipelineCheckpointOfV1, PipelineCheckpointsV1, PipelineStageInputV1, PipelineStageRunV1 } from "./case-pipeline-protocol.js";
|
|
10
|
+
export { AdmitCheckpointV1, ChallengeCheckpointV1, ControlsCheckpointV1, checkpointDigestV1, DimensionContractV1, HostFeedbackFindingKindV1, HostFeedbackFindingV1, HostFeedbackV1, makePipelineCaseContextV1, nextPipelineStageV1, OracleCheckpointV1, OracleFixtureMetadataV1, OracleScopeCoverageV1, PIPELINE_CHECKPOINT_SCHEMAS, PIPELINE_DEFAULT_FOLD_LIMIT, PIPELINE_DEFAULT_ORACLE_CONCURRENCY, PIPELINE_DEFAULT_ORACLE_REPETITIONS, PIPELINE_STAGES, PipelineBudgetExhaustedError, PipelineBudgetV1, PipelineCaseContextV1, PipelineLedgerEntryV1, PipelineLedgerPhaseV1, PipelineStageError, PipelineStageV1, PipelineStateV1, PipelineStatusV1, pipelineStageIndexV1, pipelineStageToProgressStageV1, SeedCheckpointV1, SolutionCritiqueV1, SpecCheckpointV1, SpecScopeEntryV1, TournamentCheckpointV1, TournamentHistoryEntryV1 } from "./case-pipeline-protocol.js";
|
|
11
|
+
export type { EvalCaseCursor, EvalNamedEvent, EvalNamedEventLog, EvalRunCallEvent, EvalRunFailurePhase } from "./eval-event-log.js";
|
|
12
|
+
export { assertEvalRunCanComplete, assertEvalRunCanFail, evalEventLabel, presentEvalEventLog, resumeEvalCaseCursors } from "./eval-event-log.js";
|
|
9
13
|
export { EvalRepositoryInspector, EvalRepositoryInspectorLive, inspectRepository } from "./inspection.js";
|
|
10
14
|
export { OriEvalAuthoring, oriAuthoringFromApi } from "./ori-authoring.js";
|
|
11
15
|
export type { OriEvalAuthoringApi, OriEvalResult } from "./ori-result.js";
|
|
12
16
|
export { EvalProjectArtifacts, EvalProjectArtifactsLive, evaluationProposalDigest, makeFileEvalProjectArtifacts, routingBasisDigest } from "./project-artifacts.js";
|
|
13
17
|
export type { EvalAuthoringCompletion, EvalAuthoringSource, EvalAuthoringTransportShape, EvalProjectAuthorShape } from "./project-authoring.js";
|
|
14
|
-
export { EvalAuthoringTransport, EvalProjectAuthor, EvalProjectAuthorLive, makeEvalProjectAuthor, readProjectAuthoringSources, selectProjectAuthoringSourceFiles } from "./project-authoring.js";
|
|
18
|
+
export { EVAL_AUTHORING_PREFLIGHT_OUTPUT_TOKENS, EVAL_AUTHORING_REVIEW_REQUEST_BYTES, EvalAuthoringTransport, EvalProjectAuthor, EvalProjectAuthorLive, makeEvalProjectAuthor, readProjectAuthoringSources, selectProjectAuthoringSourceFiles } from "./project-authoring.js";
|
|
15
19
|
export { EvalProjectStore, EvalProjectStoreLive, makeFileEvalProjectStore } from "./project-store.js";
|
|
16
20
|
export type { EvalProjectWorkflowError, EvalProjectWorkflowShape } from "./project-workflow.js";
|
|
17
21
|
export { EvalProjectWorkflow, EvalProjectWorkflowLive, makeEvalProjectWorkflow } from "./project-workflow.js";
|
|
18
|
-
export
|
|
19
|
-
export {
|
|
22
|
+
export { assertRepositoryAdversaryGenerationReportV1 } from "./repository-adversary-protocol.js";
|
|
23
|
+
export { assertRepositoryBehaviorMapV1 } from "./repository-behavior-protocol.js";
|
|
24
|
+
export { assertRepositoryHiddenFixtureSuiteV1 } from "./repository-fixture-protocol.js";
|
|
25
|
+
export { assertRepositoryFoundryGenerationPlanV1, assertRepositoryFoundryReferenceExclusionsV1, REPOSITORY_FOUNDRY_BASE_MODEL_CALLS, REPOSITORY_FOUNDRY_BASE_OUTPUT_CEILING, REPOSITORY_FOUNDRY_DEFAULT_WALL_TIME_MS, REPOSITORY_FOUNDRY_EXPANSIVE_OUTPUT_CEILING, REPOSITORY_FOUNDRY_HELD_OUT_REPAIR_CALLS, REPOSITORY_FOUNDRY_HILL_CLIMB_OUTPUT_CEILING, REPOSITORY_FOUNDRY_MAXIMUM_HILL_CLIMB_ATTEMPTS, REPOSITORY_FOUNDRY_MAXIMUM_VALID_CONTROL_REGENERATIONS, REPOSITORY_FOUNDRY_MAXIMUM_WALL_TIME_MS, REPOSITORY_FOUNDRY_MINIMUM_WALL_TIME_MS, REPOSITORY_FOUNDRY_PER_CALL_OUTPUT_CEILING, REPOSITORY_FOUNDRY_REQUEST_INPUT_CEILING, REPOSITORY_FOUNDRY_STANDARD_OUTPUT_CEILING, REPOSITORY_FOUNDRY_TRAJECTORY_REPAIR_ATTEMPTS, REPOSITORY_FOUNDRY_TRAJECTORY_REPAIR_CALLS, REPOSITORY_FOUNDRY_VALID_SOLUTION_REPAIR_ATTEMPTS, REPOSITORY_FOUNDRY_VALID_SOLUTION_REPAIR_CALLS, repositoryFoundryGenerationBudgetV1, repositoryFoundryGenerationPlanDigestV1, repositoryFoundryRoleOutputCeilingV1 } from "./repository-foundry-plan-protocol.js";
|
|
26
|
+
export type { RepositoryFoundryAuthoringArtifactV1, RepositoryFoundryProgressEventV1, RepositoryFoundryProgressSnapshotV1, RepositoryFoundryProgressStageV1 } from "./repository-foundry-progress-protocol.js";
|
|
27
|
+
export { boundRepositoryFoundryIdleStallMsV1, formatRepositoryFoundryProgressLineV1, REPOSITORY_FOUNDRY_DEFAULT_IDLE_STALL_MS, REPOSITORY_FOUNDRY_MAXIMUM_IDLE_STALL_MS, REPOSITORY_FOUNDRY_MINIMUM_IDLE_STALL_MS, REPOSITORY_FOUNDRY_PROGRESS_STAGES } from "./repository-foundry-progress-protocol.js";
|
|
28
|
+
export { assertRepositoryFoundryImprovementCycleV1 } from "./repository-improvement-protocol.js";
|
|
29
|
+
export { assertRepositoryFoundryModelPlanV1, gpt56RepositoryFoundryModelPlanV1, REPOSITORY_FOUNDRY_MODEL_ROLES, repositoryFoundryModelAssignmentV1, repositoryFoundryModelPlanV1 } from "./repository-language-model-protocol.js";
|
|
30
|
+
export { assertRepositoryHiddenOracleV1 } from "./repository-oracle-protocol.js";
|
|
31
|
+
export { admitRepositoryBenchmarkCaseV1, compileValidatedBenchmarkLibraryV1, evaluateRepositoryOracleAdequacyV1, repositoryBehavioralFingerprintV1 } from "./repository-quality-policy.js";
|
|
32
|
+
export type { RepositoryDimensionLabelReviewV1, RepositoryReviewedDimensionLabelV1, RepositoryRoutingBenchmarkV1, RepositoryRoutingCaseAssignmentV1, RepositoryRoutingDimensionSliceV1, RepositoryRoutingEligibilityPolicyV1 } from "./repository-routing-benchmark-protocol.js";
|
|
33
|
+
export { assertRepositoryReviewedDimensionLabelV1, assertRepositoryRoutingEligibilityPolicyV1 } from "./repository-routing-benchmark-protocol.js";
|
|
34
|
+
export type { RepositoryRoutingModelAssignmentV1, RepositoryRoutingModelCallV1, RepositoryRoutingModelPlanV1, RepositoryRoutingModelRoleV1 } from "./repository-routing-model-protocol.js";
|
|
35
|
+
export { assertRepositoryRoutingModelPlanV1, gpt56RepositoryRoutingModelPlanV1, REPOSITORY_ROUTING_MODEL_ROLES, repositoryRoutingModelAssignmentV1 } from "./repository-routing-model-protocol.js";
|
|
36
|
+
export type { RepositoryRoutingLabelingBudgetV1, RepositoryRoutingLabelingPlanV1 } from "./repository-routing-plan-protocol.js";
|
|
37
|
+
export { assertRepositoryRoutingLabelingPlanV1, REPOSITORY_ROUTING_LABEL_INPUT_CEILING, REPOSITORY_ROUTING_LABEL_OUTPUT_CEILING, REPOSITORY_ROUTING_LABEL_SEMANTIC_CALLS, REPOSITORY_ROUTING_LABEL_TRANSIENT_RETRIES, REPOSITORY_ROUTING_LABEL_WALL_TIME_MS, repositoryRoutingLabelingBudgetV1, repositoryRoutingLabelingPlanDigestV1 } from "./repository-routing-plan-protocol.js";
|
|
38
|
+
export { compileRepositoryRoutingBenchmarkV1 } from "./repository-routing-quality-policy.js";
|
|
39
|
+
export { assertRepositorySemanticCalibrationPlanV1, assertRepositorySemanticCalibrationReportV1, compileRepositorySemanticCalibrationPlanV1, compileRepositorySemanticCalibrationReportV1, executeRepositorySemanticCalibrationPlanV1, repositorySemanticCalibrationExampleDigestV1, repositorySemanticCalibrationInputDigestV1 } from "./repository-semantic-calibration.js";
|
|
40
|
+
export { REPOSITORY_SEMANTIC_CALIBRATION_MAXIMUM_EXAMPLES, REPOSITORY_SEMANTIC_CALIBRATION_MAXIMUM_REPETITIONS, REPOSITORY_SEMANTIC_CALIBRATION_POLICY_VERSION, RepositorySemanticCalibrationCountsV1, RepositorySemanticCalibrationExampleV1, RepositorySemanticCalibrationExecutableCheckV1, RepositorySemanticCalibrationInputV1, RepositorySemanticCalibrationManifestV1, RepositorySemanticCalibrationObservationV1, RepositorySemanticCalibrationOutcomeV1, RepositorySemanticCalibrationPlanV1, RepositorySemanticCalibrationReportV1, RepositorySemanticCalibrationThresholdsV1 } from "./repository-semantic-calibration-protocol.js";
|
|
41
|
+
export { assertRepositorySpecificationCritiqueV1 } from "./repository-specification-critique-protocol.js";
|
|
42
|
+
export { assertRepositoryTaskFamilyV1 } from "./repository-task-family-protocol.js";
|
|
43
|
+
export { assertRepositorySeedCrossBatchQualificationV1, assertRepositoryTaskSeedV1 } from "./repository-task-seed-protocol.js";
|
|
44
|
+
export { assertRepositoryTrajectoryPolicyV1 } from "./repository-trajectory-protocol.js";
|
|
20
45
|
export { EvalSetupRunner, EvalSetupRunnerNoop } from "./runner.js";
|
|
21
46
|
export { EvalSetup, EvalSetupLive, makeEvalSetup } from "./service.js";
|
|
47
|
+
export { Adversary, AdversaryLive, generateDeterministicAdversariesV1, makeAdversary } from "./services/adversary/service.js";
|
|
48
|
+
export type { BasisAuthoringShape } from "./services/basis-authoring/service.js";
|
|
49
|
+
export { BasisAuthoring, BasisAuthoringLive, makeBasisAuthoring } from "./services/basis-authoring/service.js";
|
|
50
|
+
export type { BasisOnboardingShape } from "./services/basis-onboarding/service.js";
|
|
51
|
+
export { BasisOnboarding, BasisOnboardingLive, makeBasisOnboarding } from "./services/basis-onboarding/service.js";
|
|
52
|
+
export type { BasisRepairShape } from "./services/basis-repair/service.js";
|
|
53
|
+
export { BasisRepair, BasisRepairLive, makeBasisRepair } from "./services/basis-repair/service.js";
|
|
54
|
+
export { BenchmarkCompiler, BenchmarkCompilerLive, compileRepositoryBenchmarkV1, makeBenchmarkCompiler } from "./services/benchmark-compiler/service.js";
|
|
55
|
+
export { authorRepositoryCaseSpecificationV1, CaseAuthoring, CaseAuthoringLive, makeCaseAuthoring } from "./services/case-authoring/service.js";
|
|
56
|
+
export { assertRepositoryFoundryFinalizationInputV1, finalizeRepositoryGeneratedCaseV1, type RepositoryFoundryFinalizationInputV1 } from "./services/case-finalization/service.js";
|
|
57
|
+
export type { RepositorySemanticReviewInputV1, RepositorySemanticReviewRoleV1, RepositorySolutionReviewProposalV1 } from "./services/case-generation/service.js";
|
|
58
|
+
export { CaseGeneration, CaseGenerationLive, executeRepositorySemanticReviewV1, generateHistoricalRepositoryCaseV1, makeCaseGeneration, prepareRepositoryCaseQualityReviewV1, prepareRepositorySemanticReviewV1, repositorySemanticReviewIndependenceFindingsV1 } from "./services/case-generation/service.js";
|
|
59
|
+
export type { CasePipelineRunInputV1, CasePipelineRunResultV1, CasePipelineStatusResultV1 } from "./services/case-pipeline/service.js";
|
|
60
|
+
export { readCasePipelineStatusV1, readRepositoryCasePipelineStatusV1, runCasePipelineV1, runRepositoryCasePipelineV1 } from "./services/case-pipeline/service.js";
|
|
61
|
+
export type { DimensionLabelingShape, RepositoryDimensionLabelingResultV1 } from "./services/dimension-labeling/service.js";
|
|
62
|
+
export { authorRepositoryReviewedDimensionLabelV1, DimensionLabeling, DimensionLabelingLive, makeDimensionLabeling } from "./services/dimension-labeling/service.js";
|
|
63
|
+
export { makeRepositoryFoundryEvidenceReconstructionV1, RepositoryFoundryEvidenceReconstruction } from "./services/evidence-reconstruction/service.js";
|
|
64
|
+
export { compileReviewedHiddenFixtureSuiteV1, FixtureBuilder, FixtureBuilderLive, makeFixtureBuilder } from "./services/fixture-builder/service.js";
|
|
65
|
+
export { executeHistoricalRepositoryCaseGenerationV1, makeRepositoryFoundry, planHistoricalRepositoryCaseGenerationV1, RepositoryFoundry, RepositoryFoundryLive, reconstructHistoricalRepositoryCaseCheckpointV1, smokeHistoricalRepositoryQualificationV1 } from "./services/foundry/service.js";
|
|
66
|
+
export type { RepositoryFoundryArtifactEmit, RepositoryFoundryProgressEmit, RepositoryFoundryProgressShape } from "./services/foundry-progress/service.js";
|
|
67
|
+
export { makeRepositoryFoundryProgress, RepositoryFoundryIdleStallError, RepositoryFoundryProgress, RepositoryFoundryProgressLive, renderRepositoryFoundryProgressLineV1, reportFoundryAuthoringArtifactV1, reportFoundryHeartbeatV1, reportFoundryModelCallV1, reportFoundryStageV1, withRepositoryFoundryIdleStall } from "./services/foundry-progress/service.js";
|
|
68
|
+
export type { HistoricalCaseShape, RepositoryHistoricalWeakOracleTrialV1, RepositoryReviewedHistoricalCaseBlueprintV1, RepositoryReviewedHistoricalCaseResultV1 } from "./services/historical-case/service.js";
|
|
69
|
+
export { compileReviewedHistoricalCaseV1, HistoricalCase, HistoricalCaseLive, makeHistoricalCase } from "./services/historical-case/service.js";
|
|
70
|
+
export { compileRepositoryFoundryImprovementCycleV1, ImprovementLoop, ImprovementLoopLive, makeImprovementLoop } from "./services/improvement-loop/service.js";
|
|
71
|
+
export type { RepositoryAssignedStructuredGenerationV1, RepositoryStructuredModelAssignmentV1 } from "./services/language-model/service.js";
|
|
72
|
+
export { makeRepositoryFoundryLanguageModel, RepositoryFoundryLanguageModel, RepositoryFoundryLanguageModelLive } from "./services/language-model/service.js";
|
|
73
|
+
export type { EvalModelCatalogShape } from "./services/model-catalog/service.js";
|
|
74
|
+
export { EvalModelCatalog } from "./services/model-catalog/service.js";
|
|
75
|
+
export { buildHistoricalHiddenOracleV1, makeOracleBuilder, OracleBuilder, OracleBuilderLive } from "./services/oracle-builder/service.js";
|
|
76
|
+
export { evaluateRepositoryBenchmarkDraftV1, makeQualityGate, QualityGate, QualityGateLive } from "./services/quality-gate/service.js";
|
|
77
|
+
export { buildRepositoryBehaviorMapV1, makeRepositoryModel, RepositoryModel, RepositoryModelLive } from "./services/repository-model/service.js";
|
|
78
|
+
export type { RepositoryRoutingLabelingExecutionResultV1, RoutingBenchmarkShape } from "./services/routing-benchmark/service.js";
|
|
79
|
+
export { executeRepositoryRoutingLabelingV1, makeRoutingBenchmark, planRepositoryRoutingLabelingV1, RoutingBenchmark, RoutingBenchmarkLive } from "./services/routing-benchmark/service.js";
|
|
80
|
+
export { critiqueRepositoryTaskSpecificationV1, makeSpecificationCritic, SpecificationCritic, SpecificationCriticLive } from "./services/specification-critic/service.js";
|
|
81
|
+
export { deriveRepositoryTaskFamiliesV1, makeTaskFamily, TaskFamily, TaskFamilyLive } from "./services/task-family/service.js";
|
|
82
|
+
export { extractHistoricalTaskSeedsV1, makeTaskSeed, qualifyHistoricalTaskSeedAcrossBatchesV1, qualifyHistoricalTaskSeedV1, TaskSeed, TaskSeedLive } from "./services/task-seed/service.js";
|
|
83
|
+
export { constructRepositoryAgentTaskV1, makeTaskSpecification, TaskSpecification, TaskSpecificationLive } from "./services/task-specification/service.js";
|
|
84
|
+
export type { RepositoryTrajectoryAuthoringResultV1, RepositoryTrajectoryPolicyProposalV1, RepositoryTrajectoryPolicyReviewV1, RepositoryTrajectoryRevisionAttemptV1, TrajectoryPolicyShape } from "./services/trajectory-policy/service.js";
|
|
85
|
+
export { authorReviewedRepositoryTrajectoryPolicyV1, makeTrajectoryPolicy, TrajectoryPolicy, TrajectoryPolicyLive } from "./services/trajectory-policy/service.js";
|
|
22
86
|
export { EvalSetupStateStore, EvalSetupStateStoreLive, initialSetupState, makeFileEvalSetupStateStore } from "./state-store.js";
|
package/dist/effect-api.js
CHANGED
|
@@ -1,15 +1,65 @@
|
|
|
1
|
-
export {
|
|
1
|
+
export { evalAuthoringRequestByteLimit, evalAuthoringResponsesRequestBody } from "./adapters/authoring-responses-request.js";
|
|
2
2
|
export { captureFrozenRepositorySnapshotV3 } from "./adapters/git-basis-snapshot.js";
|
|
3
|
-
export {
|
|
4
|
-
export {
|
|
5
|
-
export {
|
|
3
|
+
export { captureGitTreeSnapshotV1, readGitChangeEpisodesV1, readGitTreeFileV1 } from "./adapters/git-task-history.js";
|
|
4
|
+
export { inspectRepositorySolutionPathsV1, replayRepositoryCommandsV1, replayRepositoryCommandV1 } from "./adapters/repository-command-runner.js";
|
|
5
|
+
export { decodeRepositoryReviewEvidenceV1, decodeRepositoryReviewEvidenceV2, encodeRepositoryReviewEvidenceV1, encodeRepositoryReviewEvidenceV2, REPOSITORY_REVIEW_EVIDENCE_INSTRUCTIONS, REPOSITORY_REVIEW_EVIDENCE_V2_INSTRUCTIONS } from "./adapters/repository-review-evidence.js";
|
|
6
|
+
export { discoverRepositoryTestsV1 } from "./adapters/test-discovery.js";
|
|
7
|
+
export { indexTypeScriptRepositoryV1 } from "./adapters/typescript-repository-index.js";
|
|
8
|
+
export { projectRoutingBasisV3, projectRoutingDimensionV3 } from "./basis-projection.js";
|
|
9
|
+
export { AdmitCheckpointV1, ChallengeCheckpointV1, ControlsCheckpointV1, checkpointDigestV1, DimensionContractV1, HostFeedbackFindingKindV1, HostFeedbackFindingV1, HostFeedbackV1, makePipelineCaseContextV1, nextPipelineStageV1, OracleCheckpointV1, OracleFixtureMetadataV1, OracleScopeCoverageV1, PIPELINE_CHECKPOINT_SCHEMAS, PIPELINE_DEFAULT_FOLD_LIMIT, PIPELINE_DEFAULT_ORACLE_CONCURRENCY, PIPELINE_DEFAULT_ORACLE_REPETITIONS, PIPELINE_STAGES, PipelineBudgetExhaustedError, PipelineBudgetV1, PipelineCaseContextV1, PipelineLedgerEntryV1, PipelineLedgerPhaseV1, PipelineStageError, PipelineStageV1, PipelineStateV1, PipelineStatusV1, pipelineStageIndexV1, pipelineStageToProgressStageV1, SeedCheckpointV1, SolutionCritiqueV1, SpecCheckpointV1, SpecScopeEntryV1, TournamentCheckpointV1, TournamentHistoryEntryV1 } from "./case-pipeline-protocol.js";
|
|
10
|
+
export { assertEvalRunCanComplete, assertEvalRunCanFail, evalEventLabel, presentEvalEventLog, resumeEvalCaseCursors } from "./eval-event-log.js";
|
|
6
11
|
export { EvalRepositoryInspector, EvalRepositoryInspectorLive, inspectRepository } from "./inspection.js";
|
|
7
12
|
export { OriEvalAuthoring, oriAuthoringFromApi } from "./ori-authoring.js";
|
|
8
13
|
export { EvalProjectArtifacts, EvalProjectArtifactsLive, evaluationProposalDigest, makeFileEvalProjectArtifacts, routingBasisDigest } from "./project-artifacts.js";
|
|
9
|
-
export { EvalAuthoringTransport, EvalProjectAuthor, EvalProjectAuthorLive, makeEvalProjectAuthor, readProjectAuthoringSources, selectProjectAuthoringSourceFiles } from "./project-authoring.js";
|
|
14
|
+
export { EVAL_AUTHORING_PREFLIGHT_OUTPUT_TOKENS, EVAL_AUTHORING_REVIEW_REQUEST_BYTES, EvalAuthoringTransport, EvalProjectAuthor, EvalProjectAuthorLive, makeEvalProjectAuthor, readProjectAuthoringSources, selectProjectAuthoringSourceFiles } from "./project-authoring.js";
|
|
10
15
|
export { EvalProjectStore, EvalProjectStoreLive, makeFileEvalProjectStore } from "./project-store.js";
|
|
11
16
|
export { EvalProjectWorkflow, EvalProjectWorkflowLive, makeEvalProjectWorkflow } from "./project-workflow.js";
|
|
12
|
-
export {
|
|
17
|
+
export { assertRepositoryAdversaryGenerationReportV1 } from "./repository-adversary-protocol.js";
|
|
18
|
+
export { assertRepositoryBehaviorMapV1 } from "./repository-behavior-protocol.js";
|
|
19
|
+
export { assertRepositoryHiddenFixtureSuiteV1 } from "./repository-fixture-protocol.js";
|
|
20
|
+
export { assertRepositoryFoundryGenerationPlanV1, assertRepositoryFoundryReferenceExclusionsV1, REPOSITORY_FOUNDRY_BASE_MODEL_CALLS, REPOSITORY_FOUNDRY_BASE_OUTPUT_CEILING, REPOSITORY_FOUNDRY_DEFAULT_WALL_TIME_MS, REPOSITORY_FOUNDRY_EXPANSIVE_OUTPUT_CEILING, REPOSITORY_FOUNDRY_HELD_OUT_REPAIR_CALLS, REPOSITORY_FOUNDRY_HILL_CLIMB_OUTPUT_CEILING, REPOSITORY_FOUNDRY_MAXIMUM_HILL_CLIMB_ATTEMPTS, REPOSITORY_FOUNDRY_MAXIMUM_VALID_CONTROL_REGENERATIONS, REPOSITORY_FOUNDRY_MAXIMUM_WALL_TIME_MS, REPOSITORY_FOUNDRY_MINIMUM_WALL_TIME_MS, REPOSITORY_FOUNDRY_PER_CALL_OUTPUT_CEILING, REPOSITORY_FOUNDRY_REQUEST_INPUT_CEILING, REPOSITORY_FOUNDRY_STANDARD_OUTPUT_CEILING, REPOSITORY_FOUNDRY_TRAJECTORY_REPAIR_ATTEMPTS, REPOSITORY_FOUNDRY_TRAJECTORY_REPAIR_CALLS, REPOSITORY_FOUNDRY_VALID_SOLUTION_REPAIR_ATTEMPTS, REPOSITORY_FOUNDRY_VALID_SOLUTION_REPAIR_CALLS, repositoryFoundryGenerationBudgetV1, repositoryFoundryGenerationPlanDigestV1, repositoryFoundryRoleOutputCeilingV1 } from "./repository-foundry-plan-protocol.js";
|
|
21
|
+
export { boundRepositoryFoundryIdleStallMsV1, formatRepositoryFoundryProgressLineV1, REPOSITORY_FOUNDRY_DEFAULT_IDLE_STALL_MS, REPOSITORY_FOUNDRY_MAXIMUM_IDLE_STALL_MS, REPOSITORY_FOUNDRY_MINIMUM_IDLE_STALL_MS, REPOSITORY_FOUNDRY_PROGRESS_STAGES } from "./repository-foundry-progress-protocol.js";
|
|
22
|
+
export { assertRepositoryFoundryImprovementCycleV1 } from "./repository-improvement-protocol.js";
|
|
23
|
+
export { assertRepositoryFoundryModelPlanV1, gpt56RepositoryFoundryModelPlanV1, REPOSITORY_FOUNDRY_MODEL_ROLES, repositoryFoundryModelAssignmentV1, repositoryFoundryModelPlanV1 } from "./repository-language-model-protocol.js";
|
|
24
|
+
export { assertRepositoryHiddenOracleV1 } from "./repository-oracle-protocol.js";
|
|
25
|
+
export { admitRepositoryBenchmarkCaseV1, compileValidatedBenchmarkLibraryV1, evaluateRepositoryOracleAdequacyV1, repositoryBehavioralFingerprintV1 } from "./repository-quality-policy.js";
|
|
26
|
+
export { assertRepositoryReviewedDimensionLabelV1, assertRepositoryRoutingEligibilityPolicyV1 } from "./repository-routing-benchmark-protocol.js";
|
|
27
|
+
export { assertRepositoryRoutingModelPlanV1, gpt56RepositoryRoutingModelPlanV1, REPOSITORY_ROUTING_MODEL_ROLES, repositoryRoutingModelAssignmentV1 } from "./repository-routing-model-protocol.js";
|
|
28
|
+
export { assertRepositoryRoutingLabelingPlanV1, REPOSITORY_ROUTING_LABEL_INPUT_CEILING, REPOSITORY_ROUTING_LABEL_OUTPUT_CEILING, REPOSITORY_ROUTING_LABEL_SEMANTIC_CALLS, REPOSITORY_ROUTING_LABEL_TRANSIENT_RETRIES, REPOSITORY_ROUTING_LABEL_WALL_TIME_MS, repositoryRoutingLabelingBudgetV1, repositoryRoutingLabelingPlanDigestV1 } from "./repository-routing-plan-protocol.js";
|
|
29
|
+
export { compileRepositoryRoutingBenchmarkV1 } from "./repository-routing-quality-policy.js";
|
|
30
|
+
export { assertRepositorySemanticCalibrationPlanV1, assertRepositorySemanticCalibrationReportV1, compileRepositorySemanticCalibrationPlanV1, compileRepositorySemanticCalibrationReportV1, executeRepositorySemanticCalibrationPlanV1, repositorySemanticCalibrationExampleDigestV1, repositorySemanticCalibrationInputDigestV1 } from "./repository-semantic-calibration.js";
|
|
31
|
+
export { REPOSITORY_SEMANTIC_CALIBRATION_MAXIMUM_EXAMPLES, REPOSITORY_SEMANTIC_CALIBRATION_MAXIMUM_REPETITIONS, REPOSITORY_SEMANTIC_CALIBRATION_POLICY_VERSION, RepositorySemanticCalibrationCountsV1, RepositorySemanticCalibrationExampleV1, RepositorySemanticCalibrationExecutableCheckV1, RepositorySemanticCalibrationInputV1, RepositorySemanticCalibrationManifestV1, RepositorySemanticCalibrationObservationV1, RepositorySemanticCalibrationOutcomeV1, RepositorySemanticCalibrationPlanV1, RepositorySemanticCalibrationReportV1, RepositorySemanticCalibrationThresholdsV1 } from "./repository-semantic-calibration-protocol.js";
|
|
32
|
+
export { assertRepositorySpecificationCritiqueV1 } from "./repository-specification-critique-protocol.js";
|
|
33
|
+
export { assertRepositoryTaskFamilyV1 } from "./repository-task-family-protocol.js";
|
|
34
|
+
export { assertRepositorySeedCrossBatchQualificationV1, assertRepositoryTaskSeedV1 } from "./repository-task-seed-protocol.js";
|
|
35
|
+
export { assertRepositoryTrajectoryPolicyV1 } from "./repository-trajectory-protocol.js";
|
|
13
36
|
export { EvalSetupRunner, EvalSetupRunnerNoop } from "./runner.js";
|
|
14
37
|
export { EvalSetup, EvalSetupLive, makeEvalSetup } from "./service.js";
|
|
38
|
+
export { Adversary, AdversaryLive, generateDeterministicAdversariesV1, makeAdversary } from "./services/adversary/service.js";
|
|
39
|
+
export { BasisAuthoring, BasisAuthoringLive, makeBasisAuthoring } from "./services/basis-authoring/service.js";
|
|
40
|
+
export { BasisOnboarding, BasisOnboardingLive, makeBasisOnboarding } from "./services/basis-onboarding/service.js";
|
|
41
|
+
export { BasisRepair, BasisRepairLive, makeBasisRepair } from "./services/basis-repair/service.js";
|
|
42
|
+
export { BenchmarkCompiler, BenchmarkCompilerLive, compileRepositoryBenchmarkV1, makeBenchmarkCompiler } from "./services/benchmark-compiler/service.js";
|
|
43
|
+
export { authorRepositoryCaseSpecificationV1, CaseAuthoring, CaseAuthoringLive, makeCaseAuthoring } from "./services/case-authoring/service.js";
|
|
44
|
+
export { assertRepositoryFoundryFinalizationInputV1, finalizeRepositoryGeneratedCaseV1 } from "./services/case-finalization/service.js";
|
|
45
|
+
export { CaseGeneration, CaseGenerationLive, executeRepositorySemanticReviewV1, generateHistoricalRepositoryCaseV1, makeCaseGeneration, prepareRepositoryCaseQualityReviewV1, prepareRepositorySemanticReviewV1, repositorySemanticReviewIndependenceFindingsV1 } from "./services/case-generation/service.js";
|
|
46
|
+
export { readCasePipelineStatusV1, readRepositoryCasePipelineStatusV1, runCasePipelineV1, runRepositoryCasePipelineV1 } from "./services/case-pipeline/service.js";
|
|
47
|
+
export { authorRepositoryReviewedDimensionLabelV1, DimensionLabeling, DimensionLabelingLive, makeDimensionLabeling } from "./services/dimension-labeling/service.js";
|
|
48
|
+
export { makeRepositoryFoundryEvidenceReconstructionV1, RepositoryFoundryEvidenceReconstruction } from "./services/evidence-reconstruction/service.js";
|
|
49
|
+
export { compileReviewedHiddenFixtureSuiteV1, FixtureBuilder, FixtureBuilderLive, makeFixtureBuilder } from "./services/fixture-builder/service.js";
|
|
50
|
+
export { executeHistoricalRepositoryCaseGenerationV1, makeRepositoryFoundry, planHistoricalRepositoryCaseGenerationV1, RepositoryFoundry, RepositoryFoundryLive, reconstructHistoricalRepositoryCaseCheckpointV1, smokeHistoricalRepositoryQualificationV1 } from "./services/foundry/service.js";
|
|
51
|
+
export { makeRepositoryFoundryProgress, RepositoryFoundryIdleStallError, RepositoryFoundryProgress, RepositoryFoundryProgressLive, renderRepositoryFoundryProgressLineV1, reportFoundryAuthoringArtifactV1, reportFoundryHeartbeatV1, reportFoundryModelCallV1, reportFoundryStageV1, withRepositoryFoundryIdleStall } from "./services/foundry-progress/service.js";
|
|
52
|
+
export { compileReviewedHistoricalCaseV1, HistoricalCase, HistoricalCaseLive, makeHistoricalCase } from "./services/historical-case/service.js";
|
|
53
|
+
export { compileRepositoryFoundryImprovementCycleV1, ImprovementLoop, ImprovementLoopLive, makeImprovementLoop } from "./services/improvement-loop/service.js";
|
|
54
|
+
export { makeRepositoryFoundryLanguageModel, RepositoryFoundryLanguageModel, RepositoryFoundryLanguageModelLive } from "./services/language-model/service.js";
|
|
55
|
+
export { EvalModelCatalog } from "./services/model-catalog/service.js";
|
|
56
|
+
export { buildHistoricalHiddenOracleV1, makeOracleBuilder, OracleBuilder, OracleBuilderLive } from "./services/oracle-builder/service.js";
|
|
57
|
+
export { evaluateRepositoryBenchmarkDraftV1, makeQualityGate, QualityGate, QualityGateLive } from "./services/quality-gate/service.js";
|
|
58
|
+
export { buildRepositoryBehaviorMapV1, makeRepositoryModel, RepositoryModel, RepositoryModelLive } from "./services/repository-model/service.js";
|
|
59
|
+
export { executeRepositoryRoutingLabelingV1, makeRoutingBenchmark, planRepositoryRoutingLabelingV1, RoutingBenchmark, RoutingBenchmarkLive } from "./services/routing-benchmark/service.js";
|
|
60
|
+
export { critiqueRepositoryTaskSpecificationV1, makeSpecificationCritic, SpecificationCritic, SpecificationCriticLive } from "./services/specification-critic/service.js";
|
|
61
|
+
export { deriveRepositoryTaskFamiliesV1, makeTaskFamily, TaskFamily, TaskFamilyLive } from "./services/task-family/service.js";
|
|
62
|
+
export { extractHistoricalTaskSeedsV1, makeTaskSeed, qualifyHistoricalTaskSeedAcrossBatchesV1, qualifyHistoricalTaskSeedV1, TaskSeed, TaskSeedLive } from "./services/task-seed/service.js";
|
|
63
|
+
export { constructRepositoryAgentTaskV1, makeTaskSpecification, TaskSpecification, TaskSpecificationLive } from "./services/task-specification/service.js";
|
|
64
|
+
export { authorReviewedRepositoryTrajectoryPolicyV1, makeTrajectoryPolicy, TrajectoryPolicy, TrajectoryPolicyLive } from "./services/trajectory-policy/service.js";
|
|
15
65
|
export { EvalSetupStateStore, EvalSetupStateStoreLive, initialSetupState, makeFileEvalSetupStateStore } from "./state-store.js";
|
package/dist/errors.d.ts
CHANGED
|
@@ -106,6 +106,15 @@ export declare class EvalProjectAuthoringError extends EvalProjectAuthoringError
|
|
|
106
106
|
}> {
|
|
107
107
|
get message(): string;
|
|
108
108
|
}
|
|
109
|
+
declare const EvalEvaluationEvidenceFreshnessError_base: new <A extends Record<string, any> = {}>(args: import("effect/Types").VoidIfEmpty<{ readonly [P in keyof A as P extends "_tag" ? never : P]: A[P]; }>) => import("effect/Cause").YieldableError & {
|
|
110
|
+
readonly _tag: "EvalEvaluationEvidenceFreshnessError";
|
|
111
|
+
} & Readonly<A>;
|
|
112
|
+
export declare class EvalEvaluationEvidenceFreshnessError extends EvalEvaluationEvidenceFreshnessError_base<{
|
|
113
|
+
readonly detail: string;
|
|
114
|
+
readonly cause?: unknown;
|
|
115
|
+
}> {
|
|
116
|
+
get message(): string;
|
|
117
|
+
}
|
|
109
118
|
declare const EvalBasisSnapshotError_base: new <A extends Record<string, any> = {}>(args: import("effect/Types").VoidIfEmpty<{ readonly [P in keyof A as P extends "_tag" ? never : P]: A[P]; }>) => import("effect/Cause").YieldableError & {
|
|
110
119
|
readonly _tag: "EvalBasisSnapshotError";
|
|
111
120
|
} & Readonly<A>;
|
|
@@ -116,4 +125,26 @@ export declare class EvalBasisSnapshotError extends EvalBasisSnapshotError_base<
|
|
|
116
125
|
}> {
|
|
117
126
|
get message(): string;
|
|
118
127
|
}
|
|
128
|
+
declare const RepositoryFoundryError_base: new <A extends Record<string, any> = {}>(args: import("effect/Types").VoidIfEmpty<{ readonly [P in keyof A as P extends "_tag" ? never : P]: A[P]; }>) => import("effect/Cause").YieldableError & {
|
|
129
|
+
readonly _tag: "RepositoryFoundryError";
|
|
130
|
+
} & Readonly<A>;
|
|
131
|
+
export declare class RepositoryFoundryError extends RepositoryFoundryError_base<{
|
|
132
|
+
readonly operation: "resolve-repository" | "read-git-tree" | "read-git-history" | "index-repository" | "extract-task-seeds" | "select-generation-context" | "prepare-checkout" | "run-command" | "invoke-language-model" | "plan-generation" | "execute-generation" | "compile-reviewed-case" | "compile-improvement-cycle" | "run-pipeline-stage" | "checkpoint-store";
|
|
133
|
+
readonly detail: string;
|
|
134
|
+
/** Typed response-shape failure; absence does not imply a successful submission. */
|
|
135
|
+
readonly modelResponseFailure?: "invalid-json" | "normalization" | "schema";
|
|
136
|
+
/** Host decision after validated specification criticism, never provider error prose. */
|
|
137
|
+
readonly generationRejection?: {
|
|
138
|
+
readonly code: "specification-dimension-mismatch" | "specification-review-exhausted";
|
|
139
|
+
readonly completedRevisions: number;
|
|
140
|
+
};
|
|
141
|
+
/**
|
|
142
|
+
* Scientific execution classification retained independently from private
|
|
143
|
+
* command output. Environment failures mean no grade was established.
|
|
144
|
+
*/
|
|
145
|
+
readonly classification?: "environment_blocked";
|
|
146
|
+
readonly cause?: unknown;
|
|
147
|
+
}> {
|
|
148
|
+
get message(): string;
|
|
149
|
+
}
|
|
119
150
|
export {};
|
package/dist/errors.js
CHANGED
|
@@ -54,8 +54,18 @@ export class EvalProjectAuthoringError extends Data.TaggedError("EvalProjectAuth
|
|
|
54
54
|
return `RouteKit Eval could not complete ${this.operation}: ${this.detail}`;
|
|
55
55
|
}
|
|
56
56
|
}
|
|
57
|
+
export class EvalEvaluationEvidenceFreshnessError extends Data.TaggedError("EvalEvaluationEvidenceFreshnessError") {
|
|
58
|
+
get message() {
|
|
59
|
+
return `RouteKit Eval evidence is stale: ${this.detail}`;
|
|
60
|
+
}
|
|
61
|
+
}
|
|
57
62
|
export class EvalBasisSnapshotError extends Data.TaggedError("EvalBasisSnapshotError") {
|
|
58
63
|
get message() {
|
|
59
64
|
return `RouteKit Eval could not ${this.operation}: ${this.detail}`;
|
|
60
65
|
}
|
|
61
66
|
}
|
|
67
|
+
export class RepositoryFoundryError extends Data.TaggedError("RepositoryFoundryError") {
|
|
68
|
+
get message() {
|
|
69
|
+
return `RouteKit repository benchmark foundry could not ${this.operation}: ${this.detail}`;
|
|
70
|
+
}
|
|
71
|
+
}
|