agent-hitch 0.2.6 → 0.2.8
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/README.md +30 -3
- package/README.zh-CN.md +14 -2
- package/dist/bin/hitch.js +25 -3
- package/dist/bin/hitch.js.map +1 -1
- package/dist/scripts/canary-benchmark.js +54 -0
- package/dist/scripts/canary-benchmark.js.map +1 -0
- package/dist/scripts/canary-buildkit-secrets.js +166 -0
- package/dist/scripts/canary-buildkit-secrets.js.map +1 -0
- package/dist/scripts/canary-collision-domains.js +77 -0
- package/dist/scripts/canary-collision-domains.js.map +1 -0
- package/dist/scripts/canary-docker-image-observation.js +181 -0
- package/dist/scripts/canary-docker-image-observation.js.map +1 -0
- package/dist/scripts/canary-eval-scheduler-throughput.js +47 -0
- package/dist/scripts/canary-eval-scheduler-throughput.js.map +1 -0
- package/dist/scripts/canary-gpu-hardware.js +95 -0
- package/dist/scripts/canary-gpu-hardware.js.map +1 -0
- package/dist/scripts/canary-harbor-load.js +365 -0
- package/dist/scripts/canary-harbor-load.js.map +1 -0
- package/dist/scripts/canary-packaged-harness.js +75 -0
- package/dist/scripts/canary-packaged-harness.js.map +1 -0
- package/dist/scripts/canary-resource-load.js +188 -0
- package/dist/scripts/canary-resource-load.js.map +1 -0
- package/dist/scripts/check-architecture.js +10 -5
- package/dist/scripts/check-architecture.js.map +1 -1
- package/dist/scripts/docker-reaper-crash-child.js +21 -0
- package/dist/scripts/docker-reaper-crash-child.js.map +1 -0
- package/dist/src/adapters/catalog.js +3 -1
- package/dist/src/adapters/catalog.js.map +1 -1
- package/dist/src/adapters/discovery.js +1 -1
- package/dist/src/adapters/discovery.js.map +1 -1
- package/dist/src/adapters/providers/claude.js +6 -0
- package/dist/src/adapters/providers/claude.js.map +1 -1
- package/dist/src/adapters/providers/codex-auth.js +29 -0
- package/dist/src/adapters/providers/codex-auth.js.map +1 -0
- package/dist/src/adapters/providers/codex.js +29 -6
- package/dist/src/adapters/providers/codex.js.map +1 -1
- package/dist/src/adapters/providers/deepseek.js +6 -0
- package/dist/src/adapters/providers/deepseek.js.map +1 -1
- package/dist/src/adapters/providers/model-call.js +26 -0
- package/dist/src/adapters/providers/model-call.js.map +1 -0
- package/dist/src/adapters/providers/opencode.js +6 -0
- package/dist/src/adapters/providers/opencode.js.map +1 -1
- package/dist/src/adapters/providers/pi.js +6 -0
- package/dist/src/adapters/providers/pi.js.map +1 -1
- package/dist/src/artifacts/index.js +1 -0
- package/dist/src/artifacts/index.js.map +1 -1
- package/dist/src/artifacts/preparer.js +7 -2
- package/dist/src/artifacts/preparer.js.map +1 -1
- package/dist/src/backends/harbor/agent-budget.js +31 -0
- package/dist/src/backends/harbor/agent-budget.js.map +1 -0
- package/dist/src/backends/harbor/backend.js +59 -107
- package/dist/src/backends/harbor/backend.js.map +1 -1
- package/dist/src/backends/harbor/environment-config.js +92 -0
- package/dist/src/backends/harbor/environment-config.js.map +1 -0
- package/dist/src/backends/harbor/index.js +6 -1
- package/dist/src/backends/harbor/index.js.map +1 -1
- package/dist/src/backends/harbor/model-proxy-config.js +23 -0
- package/dist/src/backends/harbor/model-proxy-config.js.map +1 -0
- package/dist/src/backends/harbor/process.js +177 -0
- package/dist/src/backends/harbor/process.js.map +1 -0
- package/dist/src/backends/harbor/regrade.js +77 -0
- package/dist/src/backends/harbor/regrade.js.map +1 -0
- package/dist/src/backends/harbor/runtime-toolchain.js +30 -0
- package/dist/src/backends/harbor/runtime-toolchain.js.map +1 -0
- package/dist/src/backends/harbor/supervisor.js +74 -0
- package/dist/src/backends/harbor/supervisor.js.map +1 -0
- package/dist/src/backends/harbor/task-resources.js +206 -0
- package/dist/src/backends/harbor/task-resources.js.map +1 -0
- package/dist/src/backends/harbor/tools.js +6 -29
- package/dist/src/backends/harbor/tools.js.map +1 -1
- package/dist/src/backends/index.js +5 -1
- package/dist/src/backends/index.js.map +1 -1
- package/dist/src/benchmarks/index.js +3 -0
- package/dist/src/benchmarks/index.js.map +1 -0
- package/dist/src/benchmarks/loader.js +271 -0
- package/dist/src/benchmarks/loader.js.map +1 -0
- package/dist/src/benchmarks/metrics.js +16 -0
- package/dist/src/benchmarks/metrics.js.map +1 -0
- package/dist/src/benchmarks/toml.js +13 -0
- package/dist/src/benchmarks/toml.js.map +1 -0
- package/dist/src/benchmarks/validation.js +242 -0
- package/dist/src/benchmarks/validation.js.map +1 -0
- package/dist/src/cli/arguments.js +6 -4
- package/dist/src/cli/arguments.js.map +1 -1
- package/dist/src/cli/commands/benchmark.js +22 -0
- package/dist/src/cli/commands/benchmark.js.map +1 -0
- package/dist/src/cli/commands/capabilities.js +18 -0
- package/dist/src/cli/commands/capabilities.js.map +1 -0
- package/dist/src/cli/commands/daemon.js +135 -8
- package/dist/src/cli/commands/daemon.js.map +1 -1
- package/dist/src/cli/commands/eval.js +228 -6
- package/dist/src/cli/commands/eval.js.map +1 -1
- package/dist/src/cli/commands/images.js +40 -0
- package/dist/src/cli/commands/images.js.map +1 -0
- package/dist/src/cli/commands/run.js +42 -17
- package/dist/src/cli/commands/run.js.map +1 -1
- package/dist/src/cli/commands/runs.js +25 -3
- package/dist/src/cli/commands/runs.js.map +1 -1
- package/dist/src/cli/commands/trajectory.js +96 -5
- package/dist/src/cli/commands/trajectory.js.map +1 -1
- package/dist/src/cli/commands/verifier.js +36 -0
- package/dist/src/cli/commands/verifier.js.map +1 -0
- package/dist/src/cli/commands/worker.js +79 -0
- package/dist/src/cli/commands/worker.js.map +1 -0
- package/dist/src/cli/main.js +10 -0
- package/dist/src/cli/main.js.map +1 -1
- package/dist/src/cli/output.js +84 -4
- package/dist/src/cli/output.js.map +1 -1
- package/dist/src/control-plane/build-admission.js +69 -0
- package/dist/src/control-plane/build-admission.js.map +1 -0
- package/dist/src/control-plane/build-records.js +6 -0
- package/dist/src/control-plane/build-records.js.map +1 -0
- package/dist/src/control-plane/collisions.js +40 -0
- package/dist/src/control-plane/collisions.js.map +1 -0
- package/dist/src/control-plane/eval-control-state.js +22 -0
- package/dist/src/control-plane/eval-control-state.js.map +1 -0
- package/dist/src/control-plane/eval-control-work.js +44 -0
- package/dist/src/control-plane/eval-control-work.js.map +1 -0
- package/dist/src/control-plane/eval-image-resolution.js +52 -0
- package/dist/src/control-plane/eval-image-resolution.js.map +1 -0
- package/dist/src/control-plane/eval-image-services.js +30 -0
- package/dist/src/control-plane/eval-image-services.js.map +1 -0
- package/dist/src/control-plane/eval-records.js +252 -0
- package/dist/src/control-plane/eval-records.js.map +1 -0
- package/dist/src/control-plane/eval-recovery.js +181 -0
- package/dist/src/control-plane/eval-recovery.js.map +1 -0
- package/dist/src/control-plane/eval-scheduler.js +450 -0
- package/dist/src/control-plane/eval-scheduler.js.map +1 -0
- package/dist/src/control-plane/index.js +20 -0
- package/dist/src/control-plane/index.js.map +1 -0
- package/dist/src/control-plane/local-worker.js +49 -0
- package/dist/src/control-plane/local-worker.js.map +1 -0
- package/dist/src/control-plane/model-capture-planning.js +12 -0
- package/dist/src/control-plane/model-capture-planning.js.map +1 -0
- package/dist/src/control-plane/remote-result-transport.js +175 -0
- package/dist/src/control-plane/remote-result-transport.js.map +1 -0
- package/dist/src/control-plane/remote-work-coordinator.js +333 -0
- package/dist/src/control-plane/remote-work-coordinator.js.map +1 -0
- package/dist/src/control-plane/remote-work-inputs.js +266 -0
- package/dist/src/control-plane/remote-work-inputs.js.map +1 -0
- package/dist/src/control-plane/remote-work-item.js +62 -0
- package/dist/src/control-plane/remote-work-item.js.map +1 -0
- package/dist/src/control-plane/remote-work-recovery.js +220 -0
- package/dist/src/control-plane/remote-work-recovery.js.map +1 -0
- package/dist/src/control-plane/remote-worker-artifacts.js +135 -0
- package/dist/src/control-plane/remote-worker-artifacts.js.map +1 -0
- package/dist/src/control-plane/remote-worker-client.js +202 -0
- package/dist/src/control-plane/remote-worker-client.js.map +1 -0
- package/dist/src/control-plane/remote-worker-configuration.js +34 -0
- package/dist/src/control-plane/remote-worker-configuration.js.map +1 -0
- package/dist/src/control-plane/remote-worker-credentials.js +58 -0
- package/dist/src/control-plane/remote-worker-credentials.js.map +1 -0
- package/dist/src/control-plane/remote-worker-protocol.js +489 -0
- package/dist/src/control-plane/remote-worker-protocol.js.map +1 -0
- package/dist/src/control-plane/remote-worker-runner.js +214 -0
- package/dist/src/control-plane/remote-worker-runner.js.map +1 -0
- package/dist/src/control-plane/remote-workers.js +377 -0
- package/dist/src/control-plane/remote-workers.js.map +1 -0
- package/dist/src/control-plane/rerun-scheduler.js +448 -0
- package/dist/src/control-plane/rerun-scheduler.js.map +1 -0
- package/dist/src/control-plane/rerun-submission.js +62 -0
- package/dist/src/control-plane/rerun-submission.js.map +1 -0
- package/dist/src/control-plane/resources.js +182 -0
- package/dist/src/control-plane/resources.js.map +1 -0
- package/dist/src/control-plane/scheduler-eval-entry.js +32 -0
- package/dist/src/control-plane/scheduler-eval-entry.js.map +1 -0
- package/dist/src/control-plane/synthetic-result.js +21 -0
- package/dist/src/control-plane/synthetic-result.js.map +1 -0
- package/dist/src/control-plane/work-admission.js +23 -0
- package/dist/src/control-plane/work-admission.js.map +1 -0
- package/dist/src/control-plane/work-dispatcher.js +237 -0
- package/dist/src/control-plane/work-dispatcher.js.map +1 -0
- package/dist/src/controller-runtime/hash.js +19 -3
- package/dist/src/controller-runtime/hash.js.map +1 -1
- package/dist/src/controller-runtime/index.js +1 -1
- package/dist/src/controller-runtime/index.js.map +1 -1
- package/dist/src/controller-runtime/store.js +15 -6
- package/dist/src/controller-runtime/store.js.map +1 -1
- package/dist/src/daemon/auth.js +5 -0
- package/dist/src/daemon/auth.js.map +1 -1
- package/dist/src/daemon/client.js.map +1 -1
- package/dist/src/daemon/event-stream.js +46 -0
- package/dist/src/daemon/event-stream.js.map +1 -0
- package/dist/src/daemon/health.js +66 -0
- package/dist/src/daemon/health.js.map +1 -0
- package/dist/src/daemon/launcher.js +28 -1
- package/dist/src/daemon/launcher.js.map +1 -1
- package/dist/src/daemon/logging.js +23 -0
- package/dist/src/daemon/logging.js.map +1 -0
- package/dist/src/daemon/scheduler.js +26 -5
- package/dist/src/daemon/scheduler.js.map +1 -1
- package/dist/src/daemon/server.js +255 -98
- package/dist/src/daemon/server.js.map +1 -1
- package/dist/src/daemon/telemetry.js +163 -0
- package/dist/src/daemon/telemetry.js.map +1 -0
- package/dist/src/daemon/worker-routes.js +219 -0
- package/dist/src/daemon/worker-routes.js.map +1 -0
- package/dist/src/domain/benchmarks.js +2 -0
- package/dist/src/domain/benchmarks.js.map +1 -0
- package/dist/src/domain/bundles.js +2 -0
- package/dist/src/domain/bundles.js.map +1 -0
- package/dist/src/domain/control-plane.js +2 -0
- package/dist/src/domain/control-plane.js.map +1 -0
- package/dist/src/domain/execution-evidence.js +2 -0
- package/dist/src/domain/execution-evidence.js.map +1 -0
- package/dist/src/domain/execution-plan.js +2 -0
- package/dist/src/domain/execution-plan.js.map +1 -0
- package/dist/src/domain/images.js +2 -0
- package/dist/src/domain/images.js.map +1 -0
- package/dist/src/domain/index.js +13 -0
- package/dist/src/domain/index.js.map +1 -1
- package/dist/src/domain/interactions.js +2 -0
- package/dist/src/domain/interactions.js.map +1 -0
- package/dist/src/domain/providers.js +2 -0
- package/dist/src/domain/providers.js.map +1 -0
- package/dist/src/domain/resources.js +2 -0
- package/dist/src/domain/resources.js.map +1 -0
- package/dist/src/domain/runs.js.map +1 -1
- package/dist/src/domain/validation.js +11 -4
- package/dist/src/domain/validation.js.map +1 -1
- package/dist/src/domain/verifier-evidence-validation.js +214 -0
- package/dist/src/domain/verifier-evidence-validation.js.map +1 -0
- package/dist/src/domain/verifier-evidence.js +3 -0
- package/dist/src/domain/verifier-evidence.js.map +1 -0
- package/dist/src/domain/verifier-score-contract.js +232 -0
- package/dist/src/domain/verifier-score-contract.js.map +1 -0
- package/dist/src/domain/workers.js +2 -0
- package/dist/src/domain/workers.js.map +1 -0
- package/dist/src/evals/benchmark-adapter-manifest.js +219 -0
- package/dist/src/evals/benchmark-adapter-manifest.js.map +1 -0
- package/dist/src/evals/benchmark-candidate.js +36 -0
- package/dist/src/evals/benchmark-candidate.js.map +1 -0
- package/dist/src/evals/benchmark-run.js +166 -0
- package/dist/src/evals/benchmark-run.js.map +1 -0
- package/dist/src/evals/collect-only-rerun.js +183 -0
- package/dist/src/evals/collect-only-rerun.js.map +1 -0
- package/dist/src/evals/directory.js +30 -0
- package/dist/src/evals/directory.js.map +1 -0
- package/dist/src/evals/docker-engine-stats.js +89 -0
- package/dist/src/evals/docker-engine-stats.js.map +1 -0
- package/dist/src/evals/docker-ownership.js +51 -0
- package/dist/src/evals/docker-ownership.js.map +1 -0
- package/dist/src/evals/docker-reaper.js +144 -0
- package/dist/src/evals/docker-reaper.js.map +1 -0
- package/dist/src/evals/docker-resource-observer.js +207 -0
- package/dist/src/evals/docker-resource-observer.js.map +1 -0
- package/dist/src/evals/duration-estimator.js +126 -0
- package/dist/src/evals/duration-estimator.js.map +1 -0
- package/dist/src/evals/environment-image-planning.js +164 -0
- package/dist/src/evals/environment-image-planning.js.map +1 -0
- package/dist/src/evals/eval-artifact-planning.js +56 -0
- package/dist/src/evals/eval-artifact-planning.js.map +1 -0
- package/dist/src/evals/eval-finalization.js +10 -0
- package/dist/src/evals/eval-finalization.js.map +1 -0
- package/dist/src/evals/eval-lifecycle-events.js +29 -0
- package/dist/src/evals/eval-lifecycle-events.js.map +1 -0
- package/dist/src/evals/eval-logical-plan.js +51 -0
- package/dist/src/evals/eval-logical-plan.js.map +1 -0
- package/dist/src/evals/eval-result-builder.js +81 -0
- package/dist/src/evals/eval-result-builder.js.map +1 -0
- package/dist/src/evals/events.js +53 -5
- package/dist/src/evals/events.js.map +1 -1
- package/dist/src/evals/evolution-baseline.js +191 -0
- package/dist/src/evals/evolution-baseline.js.map +1 -0
- package/dist/src/evals/execution-evidence.js +116 -0
- package/dist/src/evals/execution-evidence.js.map +1 -0
- package/dist/src/evals/execution-leases.js +342 -0
- package/dist/src/evals/execution-leases.js.map +1 -0
- package/dist/src/evals/execution-plan-artifacts.js +72 -0
- package/dist/src/evals/execution-plan-artifacts.js.map +1 -0
- package/dist/src/evals/execution-plan-images.js +69 -0
- package/dist/src/evals/execution-plan-images.js.map +1 -0
- package/dist/src/evals/execution-plan-resources.js +174 -0
- package/dist/src/evals/execution-plan-resources.js.map +1 -0
- package/dist/src/evals/execution-plan.js +451 -0
- package/dist/src/evals/execution-plan.js.map +1 -0
- package/dist/src/evals/failure-classifier.js +61 -0
- package/dist/src/evals/failure-classifier.js.map +1 -0
- package/dist/src/evals/harbor-artifact-builder-image.js +99 -0
- package/dist/src/evals/harbor-artifact-builder-image.js.map +1 -0
- package/dist/src/evals/harbor-artifact-builder.js +358 -0
- package/dist/src/evals/harbor-artifact-builder.js.map +1 -0
- package/dist/src/evals/harbor-bridge-error.js +58 -3
- package/dist/src/evals/harbor-bridge-error.js.map +1 -1
- package/dist/src/evals/harbor-node-runtime.js +146 -0
- package/dist/src/evals/harbor-node-runtime.js.map +1 -0
- package/dist/src/evals/harbor-phase-timings.js +60 -0
- package/dist/src/evals/harbor-phase-timings.js.map +1 -0
- package/dist/src/evals/index.js +40 -2
- package/dist/src/evals/index.js.map +1 -1
- package/dist/src/evals/infrastructure-retry.js +208 -84
- package/dist/src/evals/infrastructure-retry.js.map +1 -1
- package/dist/src/evals/interaction-capture-import.js +49 -0
- package/dist/src/evals/interaction-capture-import.js.map +1 -0
- package/dist/src/evals/local-docker-provider.js +372 -0
- package/dist/src/evals/local-docker-provider.js.map +1 -0
- package/dist/src/evals/local-eval-planning.js +80 -0
- package/dist/src/evals/local-eval-planning.js.map +1 -0
- package/dist/src/evals/model-capture-plan.js +22 -0
- package/dist/src/evals/model-capture-plan.js.map +1 -0
- package/dist/src/evals/model-capture-runtime.js +114 -0
- package/dist/src/evals/model-capture-runtime.js.map +1 -0
- package/dist/src/evals/model-proxy-runtime-state.js +53 -0
- package/dist/src/evals/model-proxy-runtime-state.js.map +1 -0
- package/dist/src/evals/native-phase-evidence.js +440 -0
- package/dist/src/evals/native-phase-evidence.js.map +1 -0
- package/dist/src/evals/physical-retry-work.js +31 -0
- package/dist/src/evals/physical-retry-work.js.map +1 -0
- package/dist/src/evals/planned-execution-support.js +66 -0
- package/dist/src/evals/planned-execution-support.js.map +1 -0
- package/dist/src/evals/planned-execution.js +469 -0
- package/dist/src/evals/planned-execution.js.map +1 -0
- package/dist/src/evals/planned-progress-publisher.js +62 -0
- package/dist/src/evals/planned-progress-publisher.js.map +1 -0
- package/dist/src/evals/planned-remote-execution.js +40 -0
- package/dist/src/evals/planned-remote-execution.js.map +1 -0
- package/dist/src/evals/planned-retry-execution.js +174 -0
- package/dist/src/evals/planned-retry-execution.js.map +1 -0
- package/dist/src/evals/planned-retry-lifecycle.js +188 -0
- package/dist/src/evals/planned-retry-lifecycle.js.map +1 -0
- package/dist/src/evals/preparation-rerun.js +225 -0
- package/dist/src/evals/preparation-rerun.js.map +1 -0
- package/dist/src/evals/prepared-harness.js +76 -0
- package/dist/src/evals/prepared-harness.js.map +1 -0
- package/dist/src/evals/progress.js +52 -12
- package/dist/src/evals/progress.js.map +1 -1
- package/dist/src/evals/records.js +20 -3
- package/dist/src/evals/records.js.map +1 -1
- package/dist/src/evals/recovery.js +242 -0
- package/dist/src/evals/recovery.js.map +1 -0
- package/dist/src/evals/regrade-evidence.js +77 -0
- package/dist/src/evals/regrade-evidence.js.map +1 -0
- package/dist/src/evals/remote-infrastructure-retry.js +165 -0
- package/dist/src/evals/remote-infrastructure-retry.js.map +1 -0
- package/dist/src/evals/request.js +21 -3
- package/dist/src/evals/request.js.map +1 -1
- package/dist/src/evals/rerun-inputs.js +108 -0
- package/dist/src/evals/rerun-inputs.js.map +1 -0
- package/dist/src/evals/rerun-slots.js +4 -3
- package/dist/src/evals/rerun-slots.js.map +1 -1
- package/dist/src/evals/rerun-types.js +45 -0
- package/dist/src/evals/rerun-types.js.map +1 -0
- package/dist/src/evals/rerun.js +133 -88
- package/dist/src/evals/rerun.js.map +1 -1
- package/dist/src/evals/result-helpers.js +78 -0
- package/dist/src/evals/result-helpers.js.map +1 -0
- package/dist/src/evals/resume-state.js +36 -0
- package/dist/src/evals/resume-state.js.map +1 -0
- package/dist/src/evals/retry-backoff.js +20 -0
- package/dist/src/evals/retry-backoff.js.map +1 -0
- package/dist/src/evals/retry-state.js +195 -0
- package/dist/src/evals/retry-state.js.map +1 -0
- package/dist/src/evals/scheduler-metrics.js +95 -0
- package/dist/src/evals/scheduler-metrics.js.map +1 -0
- package/dist/src/evals/scheduler-trace-replay.js +107 -0
- package/dist/src/evals/scheduler-trace-replay.js.map +1 -0
- package/dist/src/evals/service-types.js +2 -0
- package/dist/src/evals/service-types.js.map +1 -0
- package/dist/src/evals/service.js +331 -347
- package/dist/src/evals/service.js.map +1 -1
- package/dist/src/evals/task-resources.js +209 -0
- package/dist/src/evals/task-resources.js.map +1 -0
- package/dist/src/evals/trial-environment-evidence.js +87 -0
- package/dist/src/evals/trial-environment-evidence.js.map +1 -0
- package/dist/src/evals/trial-execution-evidence.js +13 -0
- package/dist/src/evals/trial-execution-evidence.js.map +1 -0
- package/dist/src/evals/trial-import-identity.js +31 -0
- package/dist/src/evals/trial-import-identity.js.map +1 -0
- package/dist/src/evals/trial-import.js +183 -188
- package/dist/src/evals/trial-import.js.map +1 -1
- package/dist/src/evals/trial-publication-recovery.js +144 -0
- package/dist/src/evals/trial-publication-recovery.js.map +1 -0
- package/dist/src/evals/trial-publication.js +25 -0
- package/dist/src/evals/trial-publication.js.map +1 -0
- package/dist/src/evals/trial-reference-validation.js +49 -0
- package/dist/src/evals/trial-reference-validation.js.map +1 -0
- package/dist/src/evals/verifier-artifacts.js +317 -0
- package/dist/src/evals/verifier-artifacts.js.map +1 -0
- package/dist/src/evals/verifier-eligibility.js +25 -0
- package/dist/src/evals/verifier-eligibility.js.map +1 -0
- package/dist/src/evals/verifier-only-rerun.js +210 -0
- package/dist/src/evals/verifier-only-rerun.js.map +1 -0
- package/dist/src/evals/verifier-runtime.js +61 -0
- package/dist/src/evals/verifier-runtime.js.map +1 -0
- package/dist/src/evals/verifier-score-artifacts.js +119 -0
- package/dist/src/evals/verifier-score-artifacts.js.map +1 -0
- package/dist/src/evals/work-item-artifacts.js +23 -0
- package/dist/src/evals/work-item-artifacts.js.map +1 -0
- package/dist/src/foundation/config.js +17 -0
- package/dist/src/foundation/config.js.map +1 -1
- package/dist/src/foundation/contained-file.js +85 -0
- package/dist/src/foundation/contained-file.js.map +1 -0
- package/dist/src/foundation/credential-redaction.js +162 -0
- package/dist/src/foundation/credential-redaction.js.map +1 -0
- package/dist/src/foundation/environment-image-references.js +59 -0
- package/dist/src/foundation/environment-image-references.js.map +1 -0
- package/dist/src/foundation/executable.js +45 -5
- package/dist/src/foundation/executable.js.map +1 -1
- package/dist/src/foundation/fs.js +2 -1
- package/dist/src/foundation/fs.js.map +1 -1
- package/dist/src/foundation/index.js +6 -2
- package/dist/src/foundation/index.js.map +1 -1
- package/dist/src/foundation/process-identity.js +69 -0
- package/dist/src/foundation/process-identity.js.map +1 -0
- package/dist/src/images/context.js +47 -0
- package/dist/src/images/context.js.map +1 -0
- package/dist/src/images/docker-buildkit.js +137 -0
- package/dist/src/images/docker-buildkit.js.map +1 -0
- package/dist/src/images/dockerfile.js +67 -0
- package/dist/src/images/dockerfile.js.map +1 -0
- package/dist/src/images/gc.js +261 -0
- package/dist/src/images/gc.js.map +1 -0
- package/dist/src/images/index.js +10 -0
- package/dist/src/images/index.js.map +1 -0
- package/dist/src/images/manifest.js +109 -0
- package/dist/src/images/manifest.js.map +1 -0
- package/dist/src/images/ownership.js +5 -0
- package/dist/src/images/ownership.js.map +1 -0
- package/dist/src/images/records.js +32 -0
- package/dist/src/images/records.js.map +1 -0
- package/dist/src/images/registry.js +154 -0
- package/dist/src/images/registry.js.map +1 -0
- package/dist/src/images/service.js +224 -0
- package/dist/src/images/service.js.map +1 -0
- package/dist/src/model-access/capture.js +255 -0
- package/dist/src/model-access/capture.js.map +1 -0
- package/dist/src/model-access/index.js +5 -0
- package/dist/src/model-access/index.js.map +1 -0
- package/dist/src/model-access/policy.js +49 -0
- package/dist/src/model-access/policy.js.map +1 -0
- package/dist/src/model-access/proxy.js +336 -0
- package/dist/src/model-access/proxy.js.map +1 -0
- package/dist/src/model-access/records.js +167 -0
- package/dist/src/model-access/records.js.map +1 -0
- package/dist/src/runs/adapter-process.js +12 -0
- package/dist/src/runs/adapter-process.js.map +1 -0
- package/dist/src/runs/bundle.js +450 -0
- package/dist/src/runs/bundle.js.map +1 -0
- package/dist/src/runs/events.js +12 -1
- package/dist/src/runs/events.js.map +1 -1
- package/dist/src/runs/executor.js +46 -45
- package/dist/src/runs/executor.js.map +1 -1
- package/dist/src/runs/finalizer.js +17 -0
- package/dist/src/runs/finalizer.js.map +1 -1
- package/dist/src/runs/index.js +8 -1
- package/dist/src/runs/index.js.map +1 -1
- package/dist/src/runs/manifest.js +5 -5
- package/dist/src/runs/manifest.js.map +1 -1
- package/dist/src/runs/phase-bundle.js +86 -0
- package/dist/src/runs/phase-bundle.js.map +1 -0
- package/dist/src/runs/phase-cancellation.js +107 -0
- package/dist/src/runs/phase-cancellation.js.map +1 -0
- package/dist/src/runs/phase-group.js +113 -0
- package/dist/src/runs/phase-group.js.map +1 -0
- package/dist/src/runs/query.js +1 -1
- package/dist/src/runs/query.js.map +1 -1
- package/dist/src/runs/records.js +3 -0
- package/dist/src/runs/records.js.map +1 -1
- package/dist/src/runs/request.js +10 -3
- package/dist/src/runs/request.js.map +1 -1
- package/dist/src/runs/training-candidate.js +161 -0
- package/dist/src/runs/training-candidate.js.map +1 -0
- package/dist/src/runs/verifier-evidence-redaction.js +61 -0
- package/dist/src/runs/verifier-evidence-redaction.js.map +1 -0
- package/dist/src/runs/verifier-evidence.js +459 -0
- package/dist/src/runs/verifier-evidence.js.map +1 -0
- package/dist/src/runs/verifier-structured-evidence.js +83 -0
- package/dist/src/runs/verifier-structured-evidence.js.map +1 -0
- package/dist/src/trajectories/analysis.js +288 -0
- package/dist/src/trajectories/analysis.js.map +1 -0
- package/dist/src/trajectories/chunk-projection.js +148 -0
- package/dist/src/trajectories/chunk-projection.js.map +1 -0
- package/dist/src/trajectories/content-projection.js +389 -0
- package/dist/src/trajectories/content-projection.js.map +1 -0
- package/dist/src/trajectories/dsh-chunk-contract.js +176 -0
- package/dist/src/trajectories/dsh-chunk-contract.js.map +1 -0
- package/dist/src/trajectories/dsh-contract.js +284 -0
- package/dist/src/trajectories/dsh-contract.js.map +1 -0
- package/dist/src/trajectories/events-chunk-drill.js +95 -0
- package/dist/src/trajectories/events-chunk-drill.js.map +1 -0
- package/dist/src/trajectories/events-page.js +367 -0
- package/dist/src/trajectories/events-page.js.map +1 -0
- package/dist/src/trajectories/index.js +3 -0
- package/dist/src/trajectories/index.js.map +1 -1
- package/dist/src/trajectories/provider-capture.js +15 -24
- package/dist/src/trajectories/provider-capture.js.map +1 -1
- package/dist/src/trajectories/providers/deepseek.js +3 -3
- package/dist/src/trajectories/providers/deepseek.js.map +1 -1
- package/dist/src/trajectories/request-attempt.js +47 -0
- package/dist/src/trajectories/request-attempt.js.map +1 -0
- package/dist/src/trajectories/stream-reader.js +311 -0
- package/dist/src/trajectories/stream-reader.js.map +1 -0
- package/dist/src/trajectories/surface-fold.js +126 -0
- package/dist/src/trajectories/surface-fold.js.map +1 -0
- package/dist/src/workers/index.js +3 -0
- package/dist/src/workers/index.js.map +1 -0
- package/dist/src/workers/remote-harbor-work-spec.js +97 -0
- package/dist/src/workers/remote-harbor-work-spec.js.map +1 -0
- package/dist/src/workers/remote-harbor-worker.js +223 -0
- package/dist/src/workers/remote-harbor-worker.js.map +1 -0
- package/docs/schemas/benchmark-hook-request-v1.schema.json +65 -0
- package/docs/schemas/benchmark-hook-response-v1.schema.json +75 -0
- package/docs/schemas/benchmark-hook-v1.schema.json +33 -0
- package/docs/schemas/benchmark-lock-v1.schema.json +351 -0
- package/docs/schemas/benchmark-metric-v1.schema.json +37 -0
- package/docs/schemas/benchmark-package-v1.schema.json +167 -0
- package/docs/schemas/benchmark-phase-group.schema.json +41 -0
- package/docs/schemas/benchmark-profile-v1.schema.json +141 -0
- package/docs/schemas/benchmark-task-v1.schema.json +384 -0
- package/docs/schemas/benchmark-tool-result-v1.schema.json +30 -0
- package/docs/schemas/environment-build-index.schema.json +13 -0
- package/docs/schemas/environment-build-record.schema.json +25 -0
- package/docs/schemas/environment-image-manifest.schema.json +62 -0
- package/docs/schemas/error.schema.json +22 -0
- package/docs/schemas/eval-control.schema.json +36 -0
- package/docs/schemas/eval-execution-plan.schema.json +216 -0
- package/docs/schemas/eval-progress.schema.json +1 -15
- package/docs/schemas/eval-rerun-result.schema.json +58 -0
- package/docs/schemas/eval-rerun-state.schema.json +61 -0
- package/docs/schemas/eval-rerun-submission.schema.json +60 -0
- package/docs/schemas/eval-result.schema.json +32 -13
- package/docs/schemas/eval-submission-input.schema.json +14 -0
- package/docs/schemas/eval-submission.schema.json +68 -0
- package/docs/schemas/eval-trial-publication.schema.json +15 -0
- package/docs/schemas/eval-trial-reference.schema.json +129 -0
- package/docs/schemas/execution-evidence.schema.json +96 -0
- package/docs/schemas/execution-lease.schema.json +50 -0
- package/docs/schemas/execution-provider-status.schema.json +65 -0
- package/docs/schemas/execution-worker.schema.json +53 -0
- package/docs/schemas/harbor-process-exit.schema.json +14 -0
- package/docs/schemas/interaction-capture-ref.schema.json +45 -0
- package/docs/schemas/local-provider-execution.schema.json +33 -0
- package/docs/schemas/model-interaction.schema.json +41 -0
- package/docs/schemas/regrade-assessment-reference.schema.json +11 -0
- package/docs/schemas/remote-result-envelope.schema.json +32 -0
- package/docs/schemas/remote-tree-envelope.schema.json +25 -0
- package/docs/schemas/remote-work-offer.schema.json +58 -0
- package/docs/schemas/remote-work-receipt.schema.json +50 -0
- package/docs/schemas/remote-worker-artifact.schema.json +17 -0
- package/docs/schemas/remote-worker-event.schema.json +18 -0
- package/docs/schemas/remote-worker-heartbeat.schema.json +37 -0
- package/docs/schemas/remote-worker-registration.schema.json +52 -0
- package/docs/schemas/result-bundle-index.schema.json +133 -0
- package/docs/schemas/run-context.schema.json +15 -0
- package/docs/schemas/run-environment-images.schema.json +39 -0
- package/docs/schemas/run-manifest.schema.json +4 -0
- package/docs/schemas/run-request.schema.json +4 -0
- package/docs/schemas/training-data-candidate.schema.json +46 -0
- package/docs/schemas/trajectory-analysis.schema.json +258 -0
- package/docs/schemas/trajectory-events-page.schema.json +40 -0
- package/docs/schemas/verifier-assessment.schema.json +77 -0
- package/docs/schemas/verifier-evidence.schema.json +276 -0
- package/integrations/harbor/hitch_benchmark.py +252 -0
- package/integrations/harbor/hitch_candidate_recycle.py +262 -0
- package/integrations/harbor/hitch_harbor_agent.py +698 -568
- package/integrations/harbor/hitch_harbor_environment.py +469 -0
- package/integrations/harbor/hitch_harbor_task_resources.py +275 -0
- package/integrations/harbor/hitch_harbor_verifier.py +83 -2
- package/integrations/harbor/hitch_phase_supervisor.py +452 -0
- package/integrations/harbor/hitch_tool_client.mjs +91 -0
- package/integrations/model-call/cli.js +57 -0
- package/package.json +30 -6
|
@@ -0,0 +1,58 @@
|
|
|
1
|
+
{
|
|
2
|
+
"$schema": "https://json-schema.org/draft/2020-12/schema",
|
|
3
|
+
"$id": "https://agent-hitch.local/schemas/remote-work-offer.schema.json",
|
|
4
|
+
"title": "Hitch remote work offer V1",
|
|
5
|
+
"type": "object",
|
|
6
|
+
"additionalProperties": false,
|
|
7
|
+
"required": ["schema_version", "offer_id", "nonce", "generation", "worker_id", "lease", "work", "state", "issued_at", "expires_at"],
|
|
8
|
+
"properties": {
|
|
9
|
+
"schema_version": { "const": "1" },
|
|
10
|
+
"offer_id": { "type": "string", "pattern": "^offer_[a-f0-9]{32}$" },
|
|
11
|
+
"nonce": { "type": "string", "pattern": "^[a-f0-9]{64}$" },
|
|
12
|
+
"generation": { "type": "integer", "minimum": 1 },
|
|
13
|
+
"worker_id": { "type": "string", "pattern": "^worker_[a-z0-9][a-z0-9_-]{0,62}$" },
|
|
14
|
+
"lease": { "$ref": "execution-lease.schema.json" },
|
|
15
|
+
"work": { "$ref": "eval-execution-plan.schema.json#/properties/work_items/items" },
|
|
16
|
+
"inputs": {
|
|
17
|
+
"type": "array", "minItems": 1, "maxItems": 4,
|
|
18
|
+
"items": { "$ref": "#/$defs/inputRef" }
|
|
19
|
+
},
|
|
20
|
+
"state": { "enum": ["offered", "accepted", "rejected", "cancel-requested", "completed", "release-requested", "released", "expired"] },
|
|
21
|
+
"issued_at": { "type": "string", "format": "date-time" },
|
|
22
|
+
"expires_at": { "type": "string", "format": "date-time" },
|
|
23
|
+
"accepted_at": { "type": "string", "format": "date-time" },
|
|
24
|
+
"completed_at": { "type": "string", "format": "date-time" },
|
|
25
|
+
"released_at": { "type": "string", "format": "date-time" },
|
|
26
|
+
"rejection_code": { "type": "string", "minLength": 1 },
|
|
27
|
+
"accept_receipt_digest": { "$ref": "#/$defs/sha256" },
|
|
28
|
+
"terminal_receipt_digest": { "$ref": "#/$defs/sha256" },
|
|
29
|
+
"release_receipt_digest": { "$ref": "#/$defs/sha256" },
|
|
30
|
+
"terminal": {
|
|
31
|
+
"type": "object", "additionalProperties": false, "required": ["status", "artifacts", "sent_at"],
|
|
32
|
+
"properties": {
|
|
33
|
+
"status": { "enum": ["succeeded", "failed", "cancelled"] },
|
|
34
|
+
"artifacts": { "type": "array", "items": { "$ref": "#/$defs/artifactRef" } },
|
|
35
|
+
"sent_at": { "type": "string", "format": "date-time" }
|
|
36
|
+
}
|
|
37
|
+
}
|
|
38
|
+
},
|
|
39
|
+
"$defs": {
|
|
40
|
+
"sha256": { "type": "string", "pattern": "^sha256:[a-f0-9]{64}$" },
|
|
41
|
+
"artifactRef": {
|
|
42
|
+
"type": "object", "additionalProperties": false, "required": ["kind", "digest", "size"],
|
|
43
|
+
"properties": {
|
|
44
|
+
"kind": { "enum": ["result-bundle", "diagnostic"] }, "digest": { "$ref": "#/$defs/sha256" },
|
|
45
|
+
"size": { "type": "integer", "minimum": 0, "maximum": 536870912 }
|
|
46
|
+
}
|
|
47
|
+
},
|
|
48
|
+
"inputRef": {
|
|
49
|
+
"type": "object", "additionalProperties": false, "required": ["kind", "format", "digest", "size"],
|
|
50
|
+
"properties": {
|
|
51
|
+
"kind": { "enum": ["work-spec", "harness-artifact", "controller-runtime", "task-input"] },
|
|
52
|
+
"format": { "enum": ["json", "hitch-tree-v1"] },
|
|
53
|
+
"digest": { "$ref": "#/$defs/sha256" },
|
|
54
|
+
"size": { "type": "integer", "minimum": 1, "maximum": 268435456 }
|
|
55
|
+
}
|
|
56
|
+
}
|
|
57
|
+
}
|
|
58
|
+
}
|
|
@@ -0,0 +1,50 @@
|
|
|
1
|
+
{
|
|
2
|
+
"$schema": "https://json-schema.org/draft/2020-12/schema",
|
|
3
|
+
"$id": "https://agent-hitch.local/schemas/remote-work-receipt.schema.json",
|
|
4
|
+
"title": "Hitch remote work receipt V1",
|
|
5
|
+
"oneOf": [
|
|
6
|
+
{
|
|
7
|
+
"type": "object", "additionalProperties": false,
|
|
8
|
+
"required": ["schema_version", "offer_id", "nonce", "generation", "accepted", "sent_at"],
|
|
9
|
+
"properties": {
|
|
10
|
+
"schema_version": { "const": "1" }, "offer_id": { "$ref": "#/$defs/offerId" }, "nonce": { "$ref": "#/$defs/nonce" },
|
|
11
|
+
"generation": { "type": "integer", "minimum": 1 }, "accepted": { "type": "boolean" },
|
|
12
|
+
"rejection_code": { "type": "string", "minLength": 1 }, "sent_at": { "type": "string", "format": "date-time" }
|
|
13
|
+
}
|
|
14
|
+
},
|
|
15
|
+
{
|
|
16
|
+
"type": "object", "additionalProperties": false,
|
|
17
|
+
"required": ["schema_version", "offer_id", "nonce", "generation", "lease_id", "epoch", "status", "artifacts", "sent_at"],
|
|
18
|
+
"properties": {
|
|
19
|
+
"schema_version": { "const": "1" }, "offer_id": { "$ref": "#/$defs/offerId" }, "nonce": { "$ref": "#/$defs/nonce" },
|
|
20
|
+
"generation": { "type": "integer", "minimum": 1 }, "lease_id": { "$ref": "#/$defs/leaseId" }, "epoch": { "type": "integer", "minimum": 1 },
|
|
21
|
+
"status": { "enum": ["succeeded", "failed", "cancelled"] },
|
|
22
|
+
"artifacts": {
|
|
23
|
+
"type": "array", "items": {
|
|
24
|
+
"type": "object", "additionalProperties": false, "required": ["kind", "digest", "size"],
|
|
25
|
+
"properties": {
|
|
26
|
+
"kind": { "enum": ["result-bundle", "diagnostic"] }, "digest": { "$ref": "#/$defs/sha256" },
|
|
27
|
+
"size": { "type": "integer", "minimum": 0, "maximum": 536870912 }
|
|
28
|
+
}
|
|
29
|
+
}
|
|
30
|
+
},
|
|
31
|
+
"sent_at": { "type": "string", "format": "date-time" }
|
|
32
|
+
}
|
|
33
|
+
},
|
|
34
|
+
{
|
|
35
|
+
"type": "object", "additionalProperties": false,
|
|
36
|
+
"required": ["schema_version", "offer_id", "nonce", "generation", "lease_id", "epoch", "sent_at"],
|
|
37
|
+
"properties": {
|
|
38
|
+
"schema_version": { "const": "1" }, "offer_id": { "$ref": "#/$defs/offerId" }, "nonce": { "$ref": "#/$defs/nonce" },
|
|
39
|
+
"generation": { "type": "integer", "minimum": 1 }, "lease_id": { "$ref": "#/$defs/leaseId" },
|
|
40
|
+
"epoch": { "type": "integer", "minimum": 1 }, "sent_at": { "type": "string", "format": "date-time" }
|
|
41
|
+
}
|
|
42
|
+
}
|
|
43
|
+
],
|
|
44
|
+
"$defs": {
|
|
45
|
+
"offerId": { "type": "string", "pattern": "^offer_[a-f0-9]{32}$" },
|
|
46
|
+
"leaseId": { "type": "string", "pattern": "^lease_[a-f0-9]{32}$" },
|
|
47
|
+
"nonce": { "type": "string", "pattern": "^[a-f0-9]{64}$" },
|
|
48
|
+
"sha256": { "type": "string", "pattern": "^sha256:[a-f0-9]{64}$" }
|
|
49
|
+
}
|
|
50
|
+
}
|
|
@@ -0,0 +1,17 @@
|
|
|
1
|
+
{
|
|
2
|
+
"$schema": "https://json-schema.org/draft/2020-12/schema",
|
|
3
|
+
"$id": "https://agent-hitch.local/schemas/remote-worker-artifact.schema.json",
|
|
4
|
+
"title": "Hitch remote worker staged artifact V1",
|
|
5
|
+
"type": "object",
|
|
6
|
+
"additionalProperties": false,
|
|
7
|
+
"required": ["schema_version", "worker_id", "lease_id", "epoch", "digest", "size", "completed_at"],
|
|
8
|
+
"properties": {
|
|
9
|
+
"schema_version": { "const": "1" },
|
|
10
|
+
"worker_id": { "type": "string", "pattern": "^worker_[a-z0-9][a-z0-9_-]{0,62}$" },
|
|
11
|
+
"lease_id": { "type": "string", "pattern": "^lease_[a-f0-9]{32}$" },
|
|
12
|
+
"epoch": { "type": "integer", "minimum": 1 },
|
|
13
|
+
"digest": { "type": "string", "pattern": "^sha256:[a-f0-9]{64}$" },
|
|
14
|
+
"size": { "type": "integer", "minimum": 0, "maximum": 536870912 },
|
|
15
|
+
"completed_at": { "type": "string", "format": "date-time" }
|
|
16
|
+
}
|
|
17
|
+
}
|
|
@@ -0,0 +1,18 @@
|
|
|
1
|
+
{
|
|
2
|
+
"$schema": "https://json-schema.org/draft/2020-12/schema",
|
|
3
|
+
"$id": "https://agent-hitch.local/schemas/remote-worker-event.schema.json",
|
|
4
|
+
"title": "Hitch remote worker event V1",
|
|
5
|
+
"type": "object",
|
|
6
|
+
"additionalProperties": false,
|
|
7
|
+
"required": ["schema_version", "generation", "lease_id", "epoch", "sequence", "type", "sent_at"],
|
|
8
|
+
"properties": {
|
|
9
|
+
"schema_version": { "const": "1" },
|
|
10
|
+
"generation": { "type": "integer", "minimum": 1 },
|
|
11
|
+
"lease_id": { "type": "string", "pattern": "^lease_[a-f0-9]{32}$" },
|
|
12
|
+
"epoch": { "type": "integer", "minimum": 1 },
|
|
13
|
+
"sequence": { "type": "integer", "minimum": 1 },
|
|
14
|
+
"type": { "type": "string", "minLength": 1, "maxLength": 256 },
|
|
15
|
+
"payload": { "type": "object" },
|
|
16
|
+
"sent_at": { "type": "string", "format": "date-time" }
|
|
17
|
+
}
|
|
18
|
+
}
|
|
@@ -0,0 +1,37 @@
|
|
|
1
|
+
{
|
|
2
|
+
"$schema": "https://json-schema.org/draft/2020-12/schema",
|
|
3
|
+
"$id": "https://agent-hitch.local/schemas/remote-worker-heartbeat.schema.json",
|
|
4
|
+
"title": "Hitch remote worker heartbeat V1",
|
|
5
|
+
"type": "object",
|
|
6
|
+
"additionalProperties": false,
|
|
7
|
+
"required": ["schema_version", "generation", "health", "allocated", "active_leases", "sent_at"],
|
|
8
|
+
"properties": {
|
|
9
|
+
"schema_version": { "const": "1" },
|
|
10
|
+
"generation": { "type": "integer", "minimum": 1 },
|
|
11
|
+
"health": { "enum": ["healthy", "degraded", "unavailable"] },
|
|
12
|
+
"allocated": { "$ref": "#/$defs/resources" },
|
|
13
|
+
"active_leases": {
|
|
14
|
+
"type": "array",
|
|
15
|
+
"items": {
|
|
16
|
+
"type": "object", "additionalProperties": false, "required": ["lease_id", "epoch"],
|
|
17
|
+
"properties": {
|
|
18
|
+
"lease_id": { "type": "string", "pattern": "^lease_[a-f0-9]{32}$" },
|
|
19
|
+
"epoch": { "type": "integer", "minimum": 1 }
|
|
20
|
+
}
|
|
21
|
+
}
|
|
22
|
+
},
|
|
23
|
+
"sent_at": { "type": "string", "format": "date-time" }
|
|
24
|
+
},
|
|
25
|
+
"$defs": {
|
|
26
|
+
"resources": {
|
|
27
|
+
"type": "object", "additionalProperties": false,
|
|
28
|
+
"required": ["cpu_millis", "memory_bytes", "container_slots", "build_slots"],
|
|
29
|
+
"properties": {
|
|
30
|
+
"cpu_millis": { "type": "integer", "minimum": 0 }, "memory_bytes": { "type": "integer", "minimum": 0 },
|
|
31
|
+
"container_slots": { "type": "integer", "minimum": 0 }, "build_slots": { "type": "integer", "minimum": 0 },
|
|
32
|
+
"gpu_count": { "type": "integer", "minimum": 0 },
|
|
33
|
+
"ephemeral_disk_bytes": { "type": "integer", "minimum": 0 }
|
|
34
|
+
}
|
|
35
|
+
}
|
|
36
|
+
}
|
|
37
|
+
}
|
|
@@ -0,0 +1,52 @@
|
|
|
1
|
+
{
|
|
2
|
+
"$schema": "https://json-schema.org/draft/2020-12/schema",
|
|
3
|
+
"$id": "https://agent-hitch.local/schemas/remote-worker-registration.schema.json",
|
|
4
|
+
"title": "Hitch remote worker registration V1",
|
|
5
|
+
"type": "object",
|
|
6
|
+
"additionalProperties": false,
|
|
7
|
+
"required": ["schema_version", "worker_id", "provider", "collision_domain_id", "platforms", "backends", "features", "task_membership", "capacity"],
|
|
8
|
+
"properties": {
|
|
9
|
+
"schema_version": { "const": "1" },
|
|
10
|
+
"worker_id": { "$ref": "#/$defs/workerId" },
|
|
11
|
+
"provider": { "type": "string", "minLength": 1, "maxLength": 256 },
|
|
12
|
+
"collision_domain_id": { "type": "string", "minLength": 1, "maxLength": 256 },
|
|
13
|
+
"platforms": { "type": "array", "minItems": 1, "uniqueItems": true, "items": { "type": "string", "minLength": 1 } },
|
|
14
|
+
"backends": {
|
|
15
|
+
"type": "array", "minItems": 1,
|
|
16
|
+
"items": {
|
|
17
|
+
"type": "object", "additionalProperties": false, "required": ["id", "version"],
|
|
18
|
+
"properties": { "id": { "type": "string", "minLength": 1 }, "version": { "type": "string", "minLength": 1 } }
|
|
19
|
+
}
|
|
20
|
+
},
|
|
21
|
+
"features": {
|
|
22
|
+
"type": "object", "additionalProperties": false,
|
|
23
|
+
"required": ["docker", "buildkit", "model_proxy", "isolated_same_task_attempts"],
|
|
24
|
+
"properties": {
|
|
25
|
+
"docker": { "type": "boolean" }, "buildkit": { "type": "boolean" },
|
|
26
|
+
"model_proxy": { "type": "boolean" }, "isolated_same_task_attempts": { "type": "boolean" }
|
|
27
|
+
}
|
|
28
|
+
},
|
|
29
|
+
"task_membership": { "type": "array", "minItems": 1, "uniqueItems": true, "items": { "enum": ["known", "opaque"] } },
|
|
30
|
+
"capacity": {
|
|
31
|
+
"type": "object", "additionalProperties": false,
|
|
32
|
+
"required": ["total", "reserved_for_system", "allocatable"],
|
|
33
|
+
"properties": {
|
|
34
|
+
"total": { "$ref": "#/$defs/resources" }, "reserved_for_system": { "$ref": "#/$defs/resources" },
|
|
35
|
+
"allocatable": { "$ref": "#/$defs/resources" }
|
|
36
|
+
}
|
|
37
|
+
}
|
|
38
|
+
},
|
|
39
|
+
"$defs": {
|
|
40
|
+
"workerId": { "type": "string", "pattern": "^worker_[a-z0-9][a-z0-9_-]{0,62}$" },
|
|
41
|
+
"resources": {
|
|
42
|
+
"type": "object", "additionalProperties": false,
|
|
43
|
+
"required": ["cpu_millis", "memory_bytes", "container_slots", "build_slots"],
|
|
44
|
+
"properties": {
|
|
45
|
+
"cpu_millis": { "type": "integer", "minimum": 0 }, "memory_bytes": { "type": "integer", "minimum": 0 },
|
|
46
|
+
"container_slots": { "type": "integer", "minimum": 0 }, "build_slots": { "type": "integer", "minimum": 0 },
|
|
47
|
+
"gpu_count": { "type": "integer", "minimum": 0 },
|
|
48
|
+
"ephemeral_disk_bytes": { "type": "integer", "minimum": 0 }
|
|
49
|
+
}
|
|
50
|
+
}
|
|
51
|
+
}
|
|
52
|
+
}
|
|
@@ -0,0 +1,133 @@
|
|
|
1
|
+
{
|
|
2
|
+
"$schema": "https://json-schema.org/draft/2020-12/schema",
|
|
3
|
+
"$id": "https://agent-hitch.local/schemas/result-bundle-index.schema.json",
|
|
4
|
+
"title": "Hitch sealed result bundle index V1",
|
|
5
|
+
"type": "object",
|
|
6
|
+
"additionalProperties": false,
|
|
7
|
+
"required": ["schema_version", "run_id", "sealed", "context_identity", "files", "provenance", "bundle_digest", "created_at"],
|
|
8
|
+
"properties": {
|
|
9
|
+
"schema_version": { "const": "1" },
|
|
10
|
+
"run_id": { "type": "string", "pattern": "^run_[a-f0-9]{32}$" },
|
|
11
|
+
"sealed": { "const": true },
|
|
12
|
+
"context_identity": { "$ref": "#/$defs/sha256" },
|
|
13
|
+
"files": {
|
|
14
|
+
"type": "array",
|
|
15
|
+
"items": {
|
|
16
|
+
"type": "object",
|
|
17
|
+
"additionalProperties": false,
|
|
18
|
+
"required": ["role", "path", "size", "sha256"],
|
|
19
|
+
"properties": {
|
|
20
|
+
"role": {
|
|
21
|
+
"enum": [
|
|
22
|
+
"request", "resolution", "manifest", "result", "runtime-ref", "environment-manifest",
|
|
23
|
+
"execution-evidence", "control-events", "process-log", "workspace-evidence", "trajectory",
|
|
24
|
+
"provider-evidence", "verifier-evidence", "interaction-capture", "eval-publication", "diagnostic"
|
|
25
|
+
]
|
|
26
|
+
},
|
|
27
|
+
"path": { "$ref": "#/$defs/relativePath" },
|
|
28
|
+
"size": { "type": "integer", "minimum": 0 },
|
|
29
|
+
"sha256": { "$ref": "#/$defs/sha256" }
|
|
30
|
+
}
|
|
31
|
+
}
|
|
32
|
+
},
|
|
33
|
+
"environment": {
|
|
34
|
+
"type": "object",
|
|
35
|
+
"additionalProperties": false,
|
|
36
|
+
"required": ["images", "provider"],
|
|
37
|
+
"properties": {
|
|
38
|
+
"images": {
|
|
39
|
+
"type": "array",
|
|
40
|
+
"items": {
|
|
41
|
+
"type": "object",
|
|
42
|
+
"additionalProperties": false,
|
|
43
|
+
"required": ["image_id", "image_digest", "reference"],
|
|
44
|
+
"properties": {
|
|
45
|
+
"image_id": { "$ref": "#/$defs/sha256" },
|
|
46
|
+
"image_digest": { "$ref": "#/$defs/sha256" },
|
|
47
|
+
"reference": { "type": "string", "minLength": 1 }
|
|
48
|
+
}
|
|
49
|
+
}
|
|
50
|
+
},
|
|
51
|
+
"provider": { "type": "string", "minLength": 1 },
|
|
52
|
+
"worker_id": { "type": "string", "minLength": 1 },
|
|
53
|
+
"lease_id": { "type": "string", "pattern": "^lease_[a-f0-9]{32}$" }
|
|
54
|
+
}
|
|
55
|
+
},
|
|
56
|
+
"resources": {
|
|
57
|
+
"type": "object",
|
|
58
|
+
"additionalProperties": false,
|
|
59
|
+
"required": ["requested"],
|
|
60
|
+
"properties": {
|
|
61
|
+
"requested": { "$ref": "#/$defs/resources" },
|
|
62
|
+
"observed": {
|
|
63
|
+
"type": "object",
|
|
64
|
+
"additionalProperties": { "type": "integer", "minimum": 0 }
|
|
65
|
+
}
|
|
66
|
+
}
|
|
67
|
+
},
|
|
68
|
+
"interaction_ref": { "$ref": "#/$defs/relativePath" },
|
|
69
|
+
"capture": {
|
|
70
|
+
"type": "object",
|
|
71
|
+
"additionalProperties": false,
|
|
72
|
+
"required": ["mode", "required", "completeness", "interaction_count", "redaction"],
|
|
73
|
+
"properties": {
|
|
74
|
+
"mode": { "enum": ["off", "native", "proxy", "hybrid"] },
|
|
75
|
+
"required": { "type": "boolean" },
|
|
76
|
+
"completeness": { "enum": ["complete", "partial", "none"] },
|
|
77
|
+
"interaction_count": { "type": "integer", "minimum": 0 },
|
|
78
|
+
"redaction": {
|
|
79
|
+
"type": "object",
|
|
80
|
+
"additionalProperties": false,
|
|
81
|
+
"required": ["policy", "status", "rules"],
|
|
82
|
+
"properties": {
|
|
83
|
+
"policy": { "type": "string", "minLength": 1, "maxLength": 256 },
|
|
84
|
+
"status": { "enum": ["applied", "not-needed", "failed"] },
|
|
85
|
+
"rules": {
|
|
86
|
+
"type": "array",
|
|
87
|
+
"items": {
|
|
88
|
+
"type": "object",
|
|
89
|
+
"additionalProperties": false,
|
|
90
|
+
"required": ["rule_id", "count"],
|
|
91
|
+
"properties": {
|
|
92
|
+
"rule_id": { "type": "string", "minLength": 1 },
|
|
93
|
+
"count": { "type": "integer", "minimum": 1 }
|
|
94
|
+
}
|
|
95
|
+
}
|
|
96
|
+
}
|
|
97
|
+
}
|
|
98
|
+
}
|
|
99
|
+
}
|
|
100
|
+
},
|
|
101
|
+
"provenance": {
|
|
102
|
+
"type": "object",
|
|
103
|
+
"additionalProperties": false,
|
|
104
|
+
"properties": {
|
|
105
|
+
"harness_revision": { "anyOf": [{ "$ref": "#/$defs/sha256" }, { "type": "null" }] },
|
|
106
|
+
"artifact_id": { "$ref": "#/$defs/sha256" },
|
|
107
|
+
"controller_runtime_id": { "$ref": "#/$defs/sha256" },
|
|
108
|
+
"benchmark_id": { "type": "string" },
|
|
109
|
+
"benchmark_revision": { "type": "string" },
|
|
110
|
+
"verifier_identity": { "$ref": "#/$defs/sha256" }
|
|
111
|
+
}
|
|
112
|
+
},
|
|
113
|
+
"bundle_digest": { "$ref": "#/$defs/sha256" },
|
|
114
|
+
"created_at": { "type": "string", "format": "date-time" }
|
|
115
|
+
},
|
|
116
|
+
"$defs": {
|
|
117
|
+
"sha256": { "type": "string", "pattern": "^sha256:[0-9a-f]{64}$" },
|
|
118
|
+
"resources": {
|
|
119
|
+
"type": "object",
|
|
120
|
+
"additionalProperties": false,
|
|
121
|
+
"required": ["cpu_millis", "memory_bytes", "container_slots", "build_slots"],
|
|
122
|
+
"properties": {
|
|
123
|
+
"cpu_millis": { "type": "integer", "minimum": 0 },
|
|
124
|
+
"memory_bytes": { "type": "integer", "minimum": 0 },
|
|
125
|
+
"container_slots": { "type": "integer", "minimum": 0 },
|
|
126
|
+
"build_slots": { "type": "integer", "minimum": 0 },
|
|
127
|
+
"gpu_count": { "type": "integer", "minimum": 0 },
|
|
128
|
+
"ephemeral_disk_bytes": { "type": "integer", "minimum": 0 }
|
|
129
|
+
}
|
|
130
|
+
},
|
|
131
|
+
"relativePath": { "type": "string", "pattern": "^(?!/)(?!.*(?:^|/)\\.\\.(?:/|$))(?!.*\\\\).+$" }
|
|
132
|
+
}
|
|
133
|
+
}
|
|
@@ -34,6 +34,21 @@
|
|
|
34
34
|
"task_digest": { "type": "string", "pattern": "^sha256:[0-9a-f]{64}$" },
|
|
35
35
|
"verifier_identity": { "type": "string", "pattern": "^sha256:[0-9a-f]{64}$" }
|
|
36
36
|
}
|
|
37
|
+
},
|
|
38
|
+
{
|
|
39
|
+
"type": "object",
|
|
40
|
+
"additionalProperties": false,
|
|
41
|
+
"required": ["kind", "benchmark_id", "benchmark_revision", "task_id", "task_digest", "verifier_identity", "run_group_id", "phase_index"],
|
|
42
|
+
"properties": {
|
|
43
|
+
"kind": { "const": "benchmark_phase" },
|
|
44
|
+
"benchmark_id": { "type": "string", "minLength": 1 },
|
|
45
|
+
"benchmark_revision": { "type": "string", "minLength": 1, "not": { "pattern": "^[Ll][Aa][Tt][Ee][Ss][Tt]$" } },
|
|
46
|
+
"task_id": { "type": "string", "minLength": 1 },
|
|
47
|
+
"task_digest": { "type": "string", "pattern": "^sha256:[0-9a-f]{64}$" },
|
|
48
|
+
"verifier_identity": { "type": "string", "pattern": "^sha256:[0-9a-f]{64}$" },
|
|
49
|
+
"run_group_id": { "type": "string", "pattern": "^run_group_[a-f0-9]{32}$" },
|
|
50
|
+
"phase_index": { "type": "integer", "minimum": 1, "maximum": 9007199254740991 }
|
|
51
|
+
}
|
|
37
52
|
}
|
|
38
53
|
]
|
|
39
54
|
}
|
|
@@ -0,0 +1,39 @@
|
|
|
1
|
+
{
|
|
2
|
+
"$schema": "https://json-schema.org/draft/2020-12/schema",
|
|
3
|
+
"$id": "https://agent-hitch.local/schemas/run-environment-images.schema.json",
|
|
4
|
+
"title": "Hitch sealed run environment image evidence V1",
|
|
5
|
+
"type": "object",
|
|
6
|
+
"additionalProperties": false,
|
|
7
|
+
"required": ["schema_version", "task_id", "uses", "manifests"],
|
|
8
|
+
"properties": {
|
|
9
|
+
"schema_version": { "const": "1" },
|
|
10
|
+
"task_id": { "type": "string", "minLength": 1 },
|
|
11
|
+
"uses": {
|
|
12
|
+
"type": "array",
|
|
13
|
+
"minItems": 1,
|
|
14
|
+
"items": {
|
|
15
|
+
"type": "object",
|
|
16
|
+
"additionalProperties": false,
|
|
17
|
+
"required": ["task_ids", "image_id", "requested_reference", "reference", "manifest_digest", "platform", "resolution", "cache_hit"],
|
|
18
|
+
"properties": {
|
|
19
|
+
"task_ids": { "type": "array", "minItems": 1, "uniqueItems": true, "items": { "type": "string", "minLength": 1 } },
|
|
20
|
+
"image_id": { "$ref": "#/$defs/sha256" },
|
|
21
|
+
"requested_reference": { "type": "string", "minLength": 1 },
|
|
22
|
+
"reference": { "type": "string", "minLength": 1 },
|
|
23
|
+
"manifest_digest": { "$ref": "#/$defs/sha256" },
|
|
24
|
+
"platform": { "type": "string", "minLength": 1 },
|
|
25
|
+
"resolution": { "enum": ["registry", "prebuilt", "backend-build"] },
|
|
26
|
+
"cache_hit": { "type": "boolean" }
|
|
27
|
+
}
|
|
28
|
+
}
|
|
29
|
+
},
|
|
30
|
+
"manifests": {
|
|
31
|
+
"type": "array",
|
|
32
|
+
"minItems": 1,
|
|
33
|
+
"items": { "$ref": "environment-image-manifest.schema.json" }
|
|
34
|
+
}
|
|
35
|
+
},
|
|
36
|
+
"$defs": {
|
|
37
|
+
"sha256": { "type": "string", "pattern": "^sha256:[a-f0-9]{64}$" }
|
|
38
|
+
}
|
|
39
|
+
}
|
|
@@ -5,6 +5,10 @@
|
|
|
5
5
|
"type": "object",
|
|
6
6
|
"additionalProperties": true,
|
|
7
7
|
"required": ["schema_version", "run_id", "context", "status", "harness", "model", "protocol", "request_ref", "resolution_ref", "created_at"],
|
|
8
|
+
"allOf": [{
|
|
9
|
+
"if": { "properties": { "context": { "properties": { "kind": { "const": "benchmark_phase" } }, "required": ["kind"] } }, "required": ["context"] },
|
|
10
|
+
"then": { "required": ["parent"], "not": { "required": ["observation"] } }
|
|
11
|
+
}],
|
|
8
12
|
"properties": {
|
|
9
13
|
"schema_version": { "const": "1" },
|
|
10
14
|
"run_id": { "type": "string", "pattern": "^run_[a-f0-9]{32}$" },
|
|
@@ -5,6 +5,10 @@
|
|
|
5
5
|
"type": "object",
|
|
6
6
|
"additionalProperties": false,
|
|
7
7
|
"required": ["prompt"],
|
|
8
|
+
"allOf": [{
|
|
9
|
+
"if": { "properties": { "context": { "properties": { "kind": { "const": "benchmark_phase" } }, "required": ["kind"] } }, "required": ["context"] },
|
|
10
|
+
"then": { "required": ["parent"], "properties": { "defer_benchmark_observation": { "const": false } } }
|
|
11
|
+
}],
|
|
8
12
|
"oneOf": [
|
|
9
13
|
{
|
|
10
14
|
"required": ["harness_ref"],
|
|
@@ -0,0 +1,46 @@
|
|
|
1
|
+
{
|
|
2
|
+
"$schema": "https://json-schema.org/draft/2020-12/schema",
|
|
3
|
+
"$id": "https://agent-hitch.local/schemas/training-data-candidate.schema.json",
|
|
4
|
+
"title": "Hitch training-data candidate V1",
|
|
5
|
+
"type": "object",
|
|
6
|
+
"additionalProperties": false,
|
|
7
|
+
"required": [
|
|
8
|
+
"schema_version", "candidate_id", "source_bundle_digest", "run_id", "eligibility", "reasons",
|
|
9
|
+
"context_ref", "trajectory_ref", "verifier_ref", "metadata", "provenance_digest"
|
|
10
|
+
],
|
|
11
|
+
"properties": {
|
|
12
|
+
"schema_version": { "const": "1" },
|
|
13
|
+
"candidate_id": { "$ref": "#/$defs/sha256" },
|
|
14
|
+
"source_bundle_digest": { "$ref": "#/$defs/sha256" },
|
|
15
|
+
"run_id": { "type": "string", "pattern": "^run_[a-f0-9]{32}$" },
|
|
16
|
+
"eligibility": { "enum": ["eligible", "ineligible", "review-required"] },
|
|
17
|
+
"reasons": {
|
|
18
|
+
"type": "array",
|
|
19
|
+
"uniqueItems": true,
|
|
20
|
+
"items": { "type": "string", "pattern": "^[a-z0-9-]+$" }
|
|
21
|
+
},
|
|
22
|
+
"context_ref": { "type": "string", "minLength": 1, "maxLength": 4096 },
|
|
23
|
+
"trajectory_ref": { "type": "string", "minLength": 1, "maxLength": 4096 },
|
|
24
|
+
"verifier_ref": { "type": "string", "minLength": 1, "maxLength": 4096 },
|
|
25
|
+
"metadata": {
|
|
26
|
+
"type": "object",
|
|
27
|
+
"additionalProperties": false,
|
|
28
|
+
"required": [
|
|
29
|
+
"benchmark_id", "benchmark_revision", "harness_revision", "model_identity_resolved",
|
|
30
|
+
"capture_completeness", "redaction_policy"
|
|
31
|
+
],
|
|
32
|
+
"properties": {
|
|
33
|
+
"benchmark_id": { "type": "string", "minLength": 1 },
|
|
34
|
+
"benchmark_revision": { "type": "string", "minLength": 1 },
|
|
35
|
+
"harness_revision": { "$ref": "#/$defs/sha256" },
|
|
36
|
+
"model_identity_resolved": { "type": "boolean" },
|
|
37
|
+
"capture_completeness": { "enum": ["complete", "partial", "none"] },
|
|
38
|
+
"redaction_policy": { "type": "string", "minLength": 1, "maxLength": 256 }
|
|
39
|
+
}
|
|
40
|
+
},
|
|
41
|
+
"provenance_digest": { "$ref": "#/$defs/sha256" }
|
|
42
|
+
},
|
|
43
|
+
"$defs": {
|
|
44
|
+
"sha256": { "type": "string", "pattern": "^sha256:[0-9a-f]{64}$" }
|
|
45
|
+
}
|
|
46
|
+
}
|