self-bench 0.3.5 → 0.3.8
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/.dockerignore +11 -1
- package/Dockerfile +9 -9
- package/Dockerfile.sandbox +5 -4
- package/README.md +38 -21
- package/bun.lock +462 -95
- package/compose.yaml +64 -20
- package/dist/agent-feed.d.ts +8 -0
- package/dist/agent-feed.d.ts.map +1 -0
- package/dist/agent-feed.js +56 -0
- package/dist/agent-feed.js.map +1 -0
- package/dist/api/http.d.ts +18 -0
- package/dist/api/http.d.ts.map +1 -0
- package/dist/api/http.js +111 -0
- package/dist/api/http.js.map +1 -0
- package/dist/api/run-request.d.ts +37 -0
- package/dist/api/run-request.d.ts.map +1 -0
- package/dist/api/run-request.js +59 -0
- package/dist/api/run-request.js.map +1 -0
- package/dist/api/status.d.ts +4 -0
- package/dist/api/status.d.ts.map +1 -0
- package/dist/api/status.js +36 -0
- package/dist/api/status.js.map +1 -0
- package/dist/api/viewer-routes.d.ts +9 -0
- package/dist/api/viewer-routes.d.ts.map +1 -0
- package/dist/api/viewer-routes.js +98 -0
- package/dist/api/viewer-routes.js.map +1 -0
- package/dist/api-main.js +5 -1
- package/dist/api-main.js.map +1 -1
- package/dist/api.d.ts +7 -25
- package/dist/api.d.ts.map +1 -1
- package/dist/api.js +80 -148
- package/dist/api.js.map +1 -1
- package/dist/archive.d.ts +14 -0
- package/dist/archive.d.ts.map +1 -0
- package/dist/archive.js +114 -0
- package/dist/archive.js.map +1 -0
- package/dist/artifacts/common.d.ts +15 -0
- package/dist/artifacts/common.d.ts.map +1 -0
- package/dist/artifacts/common.js +58 -0
- package/dist/artifacts/common.js.map +1 -0
- package/dist/artifacts/gcs.d.ts +18 -0
- package/dist/artifacts/gcs.d.ts.map +1 -0
- package/dist/artifacts/gcs.js +178 -0
- package/dist/artifacts/gcs.js.map +1 -0
- package/dist/artifacts/local.d.ts +17 -0
- package/dist/artifacts/local.d.ts.map +1 -0
- package/dist/artifacts/local.js +205 -0
- package/dist/artifacts/local.js.map +1 -0
- package/dist/artifacts/types.d.ts +18 -0
- package/dist/artifacts/types.d.ts.map +1 -0
- package/dist/artifacts/types.js +2 -0
- package/dist/artifacts/types.js.map +1 -0
- package/dist/artifacts.d.ts +5 -28
- package/dist/artifacts.d.ts.map +1 -1
- package/dist/artifacts.js +5 -302
- package/dist/artifacts.js.map +1 -1
- package/dist/audit.d.ts.map +1 -1
- package/dist/audit.js +8 -2
- package/dist/audit.js.map +1 -1
- package/dist/auth/api-key-routes.d.ts +17 -0
- package/dist/auth/api-key-routes.d.ts.map +1 -0
- package/dist/auth/api-key-routes.js +57 -0
- package/dist/auth/api-key-routes.js.map +1 -0
- package/dist/auth/api-keys.d.ts +47 -0
- package/dist/auth/api-keys.d.ts.map +1 -0
- package/dist/auth/api-keys.js +102 -0
- package/dist/auth/api-keys.js.map +1 -0
- package/dist/auth/config.d.ts +17 -0
- package/dist/auth/config.d.ts.map +1 -0
- package/dist/auth/config.js +38 -0
- package/dist/auth/config.js.map +1 -0
- package/dist/auth/cookies.d.ts +12 -0
- package/dist/auth/cookies.d.ts.map +1 -0
- package/dist/auth/cookies.js +45 -0
- package/dist/auth/cookies.js.map +1 -0
- package/dist/auth/crypto.d.ts +19 -0
- package/dist/auth/crypto.d.ts.map +1 -0
- package/dist/auth/crypto.js +48 -0
- package/dist/auth/crypto.js.map +1 -0
- package/dist/auth/github.d.ts +37 -0
- package/dist/auth/github.d.ts.map +1 -0
- package/dist/auth/github.js +127 -0
- package/dist/auth/github.js.map +1 -0
- package/dist/auth/routes.d.ts +41 -0
- package/dist/auth/routes.d.ts.map +1 -0
- package/dist/auth/routes.js +153 -0
- package/dist/auth/routes.js.map +1 -0
- package/dist/auth/session-expired.d.ts +4 -0
- package/dist/auth/session-expired.d.ts.map +1 -0
- package/dist/auth/session-expired.js +17 -0
- package/dist/auth/session-expired.js.map +1 -0
- package/dist/auth/session.d.ts +23 -0
- package/dist/auth/session.d.ts.map +1 -0
- package/dist/auth/session.js +56 -0
- package/dist/auth/session.js.map +1 -0
- package/dist/auth/users.d.ts +49 -0
- package/dist/auth/users.d.ts.map +1 -0
- package/dist/auth/users.js +125 -0
- package/dist/auth/users.js.map +1 -0
- package/dist/build-metadata.js +1 -1
- package/dist/cli/api-client.d.ts +8 -0
- package/dist/cli/api-client.d.ts.map +1 -0
- package/dist/cli/api-client.js +71 -0
- package/dist/cli/api-client.js.map +1 -0
- package/dist/cli/associate.d.ts +2 -0
- package/dist/cli/associate.d.ts.map +1 -0
- package/dist/cli/associate.js +51 -0
- package/dist/cli/associate.js.map +1 -0
- package/dist/cli/help.d.ts +2 -0
- package/dist/cli/help.d.ts.map +1 -0
- package/dist/cli/help.js +52 -0
- package/dist/cli/help.js.map +1 -0
- package/dist/cli/main.d.ts +2 -0
- package/dist/cli/main.d.ts.map +1 -0
- package/dist/cli/main.js +55 -0
- package/dist/cli/main.js.map +1 -0
- package/dist/cli/replay.d.ts +3 -0
- package/dist/cli/replay.d.ts.map +1 -0
- package/dist/cli/replay.js +59 -0
- package/dist/cli/replay.js.map +1 -0
- package/dist/cli/repository.d.ts +6 -0
- package/dist/cli/repository.d.ts.map +1 -0
- package/dist/cli/repository.js +36 -0
- package/dist/cli/repository.js.map +1 -0
- package/dist/cli/run.d.ts +2 -0
- package/dist/cli/run.d.ts.map +1 -0
- package/dist/cli/run.js +93 -0
- package/dist/cli/run.js.map +1 -0
- package/dist/cli/stack-environment.d.ts +26 -0
- package/dist/cli/stack-environment.d.ts.map +1 -0
- package/dist/cli/stack-environment.js +86 -0
- package/dist/cli/stack-environment.js.map +1 -0
- package/dist/cli/stack.d.ts +6 -0
- package/dist/cli/stack.d.ts.map +1 -0
- package/dist/cli/stack.js +131 -0
- package/dist/cli/stack.js.map +1 -0
- package/dist/cli/values.d.ts +8 -0
- package/dist/cli/values.d.ts.map +1 -0
- package/dist/cli/values.js +30 -0
- package/dist/cli/values.js.map +1 -0
- package/dist/cli/view.d.ts +2 -0
- package/dist/cli/view.d.ts.map +1 -0
- package/dist/cli/view.js +40 -0
- package/dist/cli/view.js.map +1 -0
- package/dist/cli.js +2 -369
- package/dist/cli.js.map +1 -1
- package/dist/config.d.ts +16 -2
- package/dist/config.d.ts.map +1 -1
- package/dist/config.js +60 -22
- package/dist/config.js.map +1 -1
- package/dist/contracts/common.d.ts +22 -0
- package/dist/contracts/common.d.ts.map +1 -0
- package/dist/contracts/common.js +29 -0
- package/dist/contracts/common.js.map +1 -0
- package/dist/contracts/results.d.ts +66 -0
- package/dist/contracts/results.d.ts.map +1 -0
- package/dist/contracts/results.js +2 -0
- package/dist/contracts/results.js.map +1 -0
- package/dist/contracts/run.d.ts +145 -0
- package/dist/contracts/run.d.ts.map +1 -0
- package/dist/contracts/run.js +74 -0
- package/dist/contracts/run.js.map +1 -0
- package/dist/contracts/task.d.ts +232 -0
- package/dist/contracts/task.d.ts.map +1 -0
- package/dist/contracts/task.js +198 -0
- package/dist/contracts/task.js.map +1 -0
- package/dist/contracts/verify.d.ts +331 -0
- package/dist/contracts/verify.d.ts.map +1 -0
- package/dist/contracts/verify.js +84 -0
- package/dist/contracts/verify.js.map +1 -0
- package/dist/contracts.d.ts +5 -242
- package/dist/contracts.d.ts.map +1 -1
- package/dist/contracts.js +5 -127
- package/dist/contracts.js.map +1 -1
- package/dist/db/client.d.ts +12 -0
- package/dist/db/client.d.ts.map +1 -0
- package/dist/db/client.js +16 -0
- package/dist/db/client.js.map +1 -0
- package/dist/db/schema.d.ts +1256 -0
- package/dist/db/schema.d.ts.map +1 -0
- package/dist/db/schema.js +128 -0
- package/dist/db/schema.js.map +1 -0
- package/dist/environment.d.ts +9 -0
- package/dist/environment.d.ts.map +1 -0
- package/dist/environment.js +101 -0
- package/dist/environment.js.map +1 -0
- package/dist/evaluation/account.d.ts +30 -0
- package/dist/evaluation/account.d.ts.map +1 -0
- package/dist/evaluation/account.js +34 -0
- package/dist/evaluation/account.js.map +1 -0
- package/dist/evaluation/activities.d.ts +9 -0
- package/dist/evaluation/activities.d.ts.map +1 -0
- package/dist/evaluation/activities.js +40 -0
- package/dist/evaluation/activities.js.map +1 -0
- package/dist/evaluation/catalog-pricing.d.ts +3 -0
- package/dist/evaluation/catalog-pricing.d.ts.map +1 -0
- package/dist/evaluation/catalog-pricing.js +37 -0
- package/dist/evaluation/catalog-pricing.js.map +1 -0
- package/dist/evaluation/catalog.d.ts +17 -0
- package/dist/evaluation/catalog.d.ts.map +1 -0
- package/dist/evaluation/catalog.js +126 -0
- package/dist/evaluation/catalog.js.map +1 -0
- package/dist/evaluation/codex-login-process.d.ts +12 -0
- package/dist/evaluation/codex-login-process.d.ts.map +1 -0
- package/dist/evaluation/codex-login-process.js +121 -0
- package/dist/evaluation/codex-login-process.js.map +1 -0
- package/dist/evaluation/codex-login-routes.d.ts +5 -0
- package/dist/evaluation/codex-login-routes.d.ts.map +1 -0
- package/dist/evaluation/codex-login-routes.js +55 -0
- package/dist/evaluation/codex-login-routes.js.map +1 -0
- package/dist/evaluation/codex-login.d.ts +42 -0
- package/dist/evaluation/codex-login.d.ts.map +1 -0
- package/dist/evaluation/codex-login.js +143 -0
- package/dist/evaluation/codex-login.js.map +1 -0
- package/dist/evaluation/comparisons.d.ts +65 -0
- package/dist/evaluation/comparisons.d.ts.map +1 -0
- package/dist/evaluation/comparisons.js +161 -0
- package/dist/evaluation/comparisons.js.map +1 -0
- package/dist/evaluation/config.d.ts +43 -0
- package/dist/evaluation/config.d.ts.map +1 -0
- package/dist/evaluation/config.js +116 -0
- package/dist/evaluation/config.js.map +1 -0
- package/dist/evaluation/cost.d.ts +3 -0
- package/dist/evaluation/cost.d.ts.map +1 -0
- package/dist/evaluation/cost.js +95 -0
- package/dist/evaluation/cost.js.map +1 -0
- package/dist/evaluation/credential-execution.d.ts +10 -0
- package/dist/evaluation/credential-execution.d.ts.map +1 -0
- package/dist/evaluation/credential-execution.js +65 -0
- package/dist/evaluation/credential-execution.js.map +1 -0
- package/dist/evaluation/credentials.d.ts +40 -0
- package/dist/evaluation/credentials.d.ts.map +1 -0
- package/dist/evaluation/credentials.js +146 -0
- package/dist/evaluation/credentials.js.map +1 -0
- package/dist/evaluation/encrypted-records.d.ts +15 -0
- package/dist/evaluation/encrypted-records.d.ts.map +1 -0
- package/dist/evaluation/encrypted-records.js +57 -0
- package/dist/evaluation/encrypted-records.js.map +1 -0
- package/dist/evaluation/gateway-execution.d.ts +18 -0
- package/dist/evaluation/gateway-execution.d.ts.map +1 -0
- package/dist/evaluation/gateway-execution.js +38 -0
- package/dist/evaluation/gateway-execution.js.map +1 -0
- package/dist/evaluation/harbor-cost.d.ts +3 -0
- package/dist/evaluation/harbor-cost.d.ts.map +1 -0
- package/dist/evaluation/harbor-cost.js +39 -0
- package/dist/evaluation/harbor-cost.js.map +1 -0
- package/dist/evaluation/harbor_e2b.py +20 -0
- package/dist/evaluation/harbor_gateway.py +22 -0
- package/dist/evaluation/harnesses.d.ts +8 -0
- package/dist/evaluation/harnesses.d.ts.map +1 -0
- package/dist/evaluation/harnesses.js +10 -0
- package/dist/evaluation/harnesses.js.map +1 -0
- package/dist/evaluation/model-options.d.ts +9 -0
- package/dist/evaluation/model-options.d.ts.map +1 -0
- package/dist/evaluation/model-options.js +81 -0
- package/dist/evaluation/model-options.js.map +1 -0
- package/dist/evaluation/org-credential-routes.d.ts +5 -0
- package/dist/evaluation/org-credential-routes.d.ts.map +1 -0
- package/dist/evaluation/org-credential-routes.js +56 -0
- package/dist/evaluation/org-credential-routes.js.map +1 -0
- package/dist/evaluation/org-records.d.ts +4 -0
- package/dist/evaluation/org-records.d.ts.map +1 -0
- package/dist/evaluation/org-records.js +14 -0
- package/dist/evaluation/org-records.js.map +1 -0
- package/dist/evaluation/output.d.ts +9 -0
- package/dist/evaluation/output.d.ts.map +1 -0
- package/dist/evaluation/output.js +171 -0
- package/dist/evaluation/output.js.map +1 -0
- package/dist/evaluation/platform-routes.d.ts +5 -0
- package/dist/evaluation/platform-routes.d.ts.map +1 -0
- package/dist/evaluation/platform-routes.js +123 -0
- package/dist/evaluation/platform-routes.js.map +1 -0
- package/dist/evaluation/profile-routes.d.ts +16 -0
- package/dist/evaluation/profile-routes.d.ts.map +1 -0
- package/dist/evaluation/profile-routes.js +47 -0
- package/dist/evaluation/profile-routes.js.map +1 -0
- package/dist/evaluation/profiles.d.ts +48 -0
- package/dist/evaluation/profiles.d.ts.map +1 -0
- package/dist/evaluation/profiles.js +134 -0
- package/dist/evaluation/profiles.js.map +1 -0
- package/dist/evaluation/providers.d.ts +11 -0
- package/dist/evaluation/providers.d.ts.map +1 -0
- package/dist/evaluation/providers.js +30 -0
- package/dist/evaluation/providers.js.map +1 -0
- package/dist/evaluation/routes.d.ts +23 -0
- package/dist/evaluation/routes.d.ts.map +1 -0
- package/dist/evaluation/routes.js +212 -0
- package/dist/evaluation/routes.js.map +1 -0
- package/dist/evaluation/runner.d.ts +16 -0
- package/dist/evaluation/runner.d.ts.map +1 -0
- package/dist/evaluation/runner.js +241 -0
- package/dist/evaluation/runner.js.map +1 -0
- package/dist/evaluation/store.d.ts +8 -0
- package/dist/evaluation/store.d.ts.map +1 -0
- package/dist/evaluation/store.js +60 -0
- package/dist/evaluation/store.js.map +1 -0
- package/dist/evaluation/types.d.ts +98 -0
- package/dist/evaluation/types.d.ts.map +1 -0
- package/dist/evaluation/types.js +2 -0
- package/dist/evaluation/types.js.map +1 -0
- package/dist/evaluation/worker-main.d.ts +2 -0
- package/dist/evaluation/worker-main.d.ts.map +1 -0
- package/dist/evaluation/worker-main.js +26 -0
- package/dist/evaluation/worker-main.js.map +1 -0
- package/dist/evaluation/worker-records.d.ts +5 -0
- package/dist/evaluation/worker-records.d.ts.map +1 -0
- package/dist/evaluation/worker-records.js +11 -0
- package/dist/evaluation/worker-records.js.map +1 -0
- package/dist/evaluation/workflow.d.ts +3 -0
- package/dist/evaluation/workflow.d.ts.map +1 -0
- package/dist/evaluation/workflow.js +20 -0
- package/dist/evaluation/workflow.js.map +1 -0
- package/dist/execution-environment.d.ts +3 -0
- package/dist/execution-environment.d.ts.map +1 -0
- package/dist/execution-environment.js +9 -0
- package/dist/execution-environment.js.map +1 -0
- package/dist/extension-authoring.bundle.js +347 -0
- package/dist/extension-verifier.bundle.js +73 -0
- package/dist/extensions/authoring.d.ts.map +1 -1
- package/dist/extensions/authoring.js +70 -59
- package/dist/extensions/authoring.js.map +1 -1
- package/dist/extensions/discovery.d.ts.map +1 -1
- package/dist/extensions/discovery.js +12 -5
- package/dist/extensions/discovery.js.map +1 -1
- package/dist/extensions/shared/deliverable.d.ts +33 -0
- package/dist/extensions/shared/deliverable.d.ts.map +1 -0
- package/dist/extensions/shared/deliverable.js +123 -0
- package/dist/extensions/shared/deliverable.js.map +1 -0
- package/dist/extensions/shared/mailbox.d.ts +33 -0
- package/dist/extensions/shared/mailbox.d.ts.map +1 -0
- package/dist/extensions/shared/mailbox.js +102 -0
- package/dist/extensions/shared/mailbox.js.map +1 -0
- package/dist/extensions/shared/schemas.d.ts +90 -0
- package/dist/extensions/shared/schemas.d.ts.map +1 -0
- package/dist/extensions/shared/schemas.js +71 -0
- package/dist/extensions/shared/schemas.js.map +1 -0
- package/dist/extensions/shared/static-check.d.ts +38 -0
- package/dist/extensions/shared/static-check.d.ts.map +1 -0
- package/dist/extensions/shared/static-check.js +52 -0
- package/dist/extensions/shared/static-check.js.map +1 -0
- package/dist/extensions/verifier.d.ts +4 -0
- package/dist/extensions/verifier.d.ts.map +1 -0
- package/dist/extensions/verifier.js +70 -0
- package/dist/extensions/verifier.js.map +1 -0
- package/dist/github.d.ts +1 -1
- package/dist/github.d.ts.map +1 -1
- package/dist/github.js +3 -3
- package/dist/github.js.map +1 -1
- package/dist/harbor-environment.d.ts +13 -1
- package/dist/harbor-environment.d.ts.map +1 -1
- package/dist/harbor-environment.js +37 -4
- package/dist/harbor-environment.js.map +1 -1
- package/dist/harbor-results.d.ts +2 -0
- package/dist/harbor-results.d.ts.map +1 -1
- package/dist/harbor-results.js +2 -0
- package/dist/harbor-results.js.map +1 -1
- package/dist/harbor-task/compiler.d.ts +10 -0
- package/dist/harbor-task/compiler.d.ts.map +1 -0
- package/dist/harbor-task/compiler.js +174 -0
- package/dist/harbor-task/compiler.js.map +1 -0
- package/dist/harbor-task/constants.d.ts +3 -0
- package/dist/harbor-task/constants.d.ts.map +1 -0
- package/dist/harbor-task/constants.js +3 -0
- package/dist/harbor-task/constants.js.map +1 -0
- package/dist/harbor-task/dependencies.d.ts +3 -0
- package/dist/harbor-task/dependencies.d.ts.map +1 -0
- package/dist/harbor-task/dependencies.js +20 -0
- package/dist/harbor-task/dependencies.js.map +1 -0
- package/dist/harbor-task/paths.d.ts +6 -0
- package/dist/harbor-task/paths.d.ts.map +1 -0
- package/dist/harbor-task/paths.js +37 -0
- package/dist/harbor-task/paths.js.map +1 -0
- package/dist/harbor-task/render.d.ts +11 -0
- package/dist/harbor-task/render.d.ts.map +1 -0
- package/dist/harbor-task/render.js +194 -0
- package/dist/harbor-task/render.js.map +1 -0
- package/dist/harbor-task/runtime/command.sh +26 -0
- package/dist/harbor-task/runtime/junit.py +69 -0
- package/dist/harbor-task/runtime-assets.d.ts +2 -0
- package/dist/harbor-task/runtime-assets.d.ts.map +1 -0
- package/dist/harbor-task/runtime-assets.js +8 -0
- package/dist/harbor-task/runtime-assets.js.map +1 -0
- package/dist/harbor-task/test-patch.d.ts +4 -0
- package/dist/harbor-task/test-patch.d.ts.map +1 -0
- package/dist/harbor-task/test-patch.js +15 -0
- package/dist/harbor-task/test-patch.js.map +1 -0
- package/dist/harbor-task/verifier.d.ts +4 -0
- package/dist/harbor-task/verifier.d.ts.map +1 -0
- package/dist/harbor-task/verifier.js +105 -0
- package/dist/harbor-task/verifier.js.map +1 -0
- package/dist/harbor-task.d.ts +3 -11
- package/dist/harbor-task.d.ts.map +1 -1
- package/dist/harbor-task.js +2 -461
- package/dist/harbor-task.js.map +1 -1
- package/dist/log-excerpt.d.ts +15 -0
- package/dist/log-excerpt.d.ts.map +1 -0
- package/dist/log-excerpt.js +104 -0
- package/dist/log-excerpt.js.map +1 -0
- package/dist/patch-check.d.ts +26 -0
- package/dist/patch-check.d.ts.map +1 -0
- package/dist/patch-check.js +137 -0
- package/dist/patch-check.js.map +1 -0
- package/dist/pi-event-feed.d.ts +16 -0
- package/dist/pi-event-feed.d.ts.map +1 -0
- package/dist/pi-event-feed.js +86 -0
- package/dist/pi-event-feed.js.map +1 -0
- package/dist/pi-session.d.ts +32 -0
- package/dist/pi-session.d.ts.map +1 -0
- package/dist/pi-session.js +175 -0
- package/dist/pi-session.js.map +1 -0
- package/dist/process.d.ts +5 -0
- package/dist/process.d.ts.map +1 -1
- package/dist/process.js +31 -9
- package/dist/process.js.map +1 -1
- package/dist/project-paths.d.ts +2 -0
- package/dist/project-paths.d.ts.map +1 -0
- package/dist/project-paths.js +15 -0
- package/dist/project-paths.js.map +1 -0
- package/dist/provenance/github.d.ts +4 -0
- package/dist/provenance/github.d.ts.map +1 -0
- package/dist/provenance/github.js +65 -0
- package/dist/provenance/github.js.map +1 -0
- package/dist/provenance/local.d.ts +4 -0
- package/dist/provenance/local.d.ts.map +1 -0
- package/dist/provenance/local.js +82 -0
- package/dist/provenance/local.js.map +1 -0
- package/dist/provenance/redact.d.ts +2 -0
- package/dist/provenance/redact.d.ts.map +1 -0
- package/dist/provenance/redact.js +28 -0
- package/dist/provenance/redact.js.map +1 -0
- package/dist/provenance/selection.d.ts +4 -0
- package/dist/provenance/selection.d.ts.map +1 -0
- package/dist/provenance/selection.js +29 -0
- package/dist/provenance/selection.js.map +1 -0
- package/dist/provenance/session.d.ts +3 -0
- package/dist/provenance/session.d.ts.map +1 -0
- package/dist/provenance/session.js +214 -0
- package/dist/provenance/session.js.map +1 -0
- package/dist/provenance/shared.d.ts +7 -0
- package/dist/provenance/shared.d.ts.map +1 -0
- package/dist/provenance/shared.js +10 -0
- package/dist/provenance/shared.js.map +1 -0
- package/dist/provenance/types.d.ts +34 -0
- package/dist/provenance/types.d.ts.map +1 -0
- package/dist/provenance/types.js +22 -0
- package/dist/provenance/types.js.map +1 -0
- package/dist/provenance-associations/apply.d.ts +3 -0
- package/dist/provenance-associations/apply.d.ts.map +1 -0
- package/dist/provenance-associations/apply.js +66 -0
- package/dist/provenance-associations/apply.js.map +1 -0
- package/dist/provenance-associations/shared.d.ts +45 -0
- package/dist/provenance-associations/shared.d.ts.map +1 -0
- package/dist/provenance-associations/shared.js +42 -0
- package/dist/provenance-associations/shared.js.map +1 -0
- package/dist/provenance-associations.d.ts +22 -0
- package/dist/provenance-associations.d.ts.map +1 -0
- package/dist/provenance-associations.js +121 -0
- package/dist/provenance-associations.js.map +1 -0
- package/dist/provenance.d.ts +7 -20
- package/dist/provenance.d.ts.map +1 -1
- package/dist/provenance.js +6 -360
- package/dist/provenance.js.map +1 -1
- package/dist/providers.d.ts +25 -2
- package/dist/providers.d.ts.map +1 -1
- package/dist/providers.js +23 -2
- package/dist/providers.js.map +1 -1
- package/dist/repair.d.ts +1 -6
- package/dist/repair.d.ts.map +1 -1
- package/dist/repair.js +17 -27
- package/dist/repair.js.map +1 -1
- package/dist/review/assets/index-BHHztQ9x.js +1679 -0
- package/dist/review/assets/index-KQ1eFQ0t.css +1 -0
- package/dist/review/index.html +11 -8
- package/dist/runtime/command.sh +26 -0
- package/dist/runtime/junit.py +69 -0
- package/dist/sandbox/contracts.d.ts +76 -0
- package/dist/sandbox/contracts.d.ts.map +1 -0
- package/dist/sandbox/contracts.js +12 -0
- package/dist/sandbox/contracts.js.map +1 -0
- package/dist/sandbox/index.d.ts +5 -0
- package/dist/sandbox/index.d.ts.map +1 -0
- package/dist/sandbox/index.js +27 -0
- package/dist/sandbox/index.js.map +1 -0
- package/dist/sandbox/live.d.ts +18 -0
- package/dist/sandbox/live.d.ts.map +1 -0
- package/dist/sandbox/live.js +55 -0
- package/dist/sandbox/live.js.map +1 -0
- package/dist/sandbox/output-retry.d.ts +16 -0
- package/dist/sandbox/output-retry.d.ts.map +1 -0
- package/dist/sandbox/output-retry.js +36 -0
- package/dist/sandbox/output-retry.js.map +1 -0
- package/dist/sandbox/programs/author.d.ts +3 -0
- package/dist/sandbox/programs/author.d.ts.map +1 -0
- package/dist/sandbox/programs/author.js +27 -0
- package/dist/sandbox/programs/author.js.map +1 -0
- package/dist/sandbox/programs/check.d.ts +3 -0
- package/dist/sandbox/programs/check.d.ts.map +1 -0
- package/dist/sandbox/programs/check.js +76 -0
- package/dist/sandbox/programs/check.js.map +1 -0
- package/dist/sandbox/programs/prepare-task.d.ts +12 -0
- package/dist/sandbox/programs/prepare-task.d.ts.map +1 -0
- package/dist/sandbox/programs/prepare-task.js +32 -0
- package/dist/sandbox/programs/prepare-task.js.map +1 -0
- package/dist/sandbox/programs/verifier.d.ts +3 -0
- package/dist/sandbox/programs/verifier.d.ts.map +1 -0
- package/dist/sandbox/programs/verifier.js +9 -0
- package/dist/sandbox/programs/verifier.js.map +1 -0
- package/dist/sandbox/providers/docker/executor.d.ts +14 -0
- package/dist/sandbox/providers/docker/executor.d.ts.map +1 -0
- package/dist/sandbox/providers/docker/executor.js +153 -0
- package/dist/sandbox/providers/docker/executor.js.map +1 -0
- package/dist/sandbox/providers/e2b/bytes.d.ts +3 -0
- package/dist/sandbox/providers/e2b/bytes.d.ts.map +1 -0
- package/dist/sandbox/providers/e2b/bytes.js +8 -0
- package/dist/sandbox/providers/e2b/bytes.js.map +1 -0
- package/dist/sandbox/providers/e2b/cleanup.d.ts +8 -0
- package/dist/sandbox/providers/e2b/cleanup.d.ts.map +1 -0
- package/dist/sandbox/providers/e2b/cleanup.js +124 -0
- package/dist/sandbox/providers/e2b/cleanup.js.map +1 -0
- package/dist/sandbox/providers/e2b/command.d.ts +23 -0
- package/dist/sandbox/providers/e2b/command.d.ts.map +1 -0
- package/dist/sandbox/providers/e2b/command.js +160 -0
- package/dist/sandbox/providers/e2b/command.js.map +1 -0
- package/dist/sandbox/providers/e2b/config.d.ts +13 -0
- package/dist/sandbox/providers/e2b/config.d.ts.map +1 -0
- package/dist/sandbox/providers/e2b/config.js +2 -0
- package/dist/sandbox/providers/e2b/config.js.map +1 -0
- package/dist/sandbox/providers/e2b/executor.d.ts +15 -0
- package/dist/sandbox/providers/e2b/executor.d.ts.map +1 -0
- package/dist/sandbox/providers/e2b/executor.js +207 -0
- package/dist/sandbox/providers/e2b/executor.js.map +1 -0
- package/dist/sandbox/providers/e2b/lifecycle.d.ts +8 -0
- package/dist/sandbox/providers/e2b/lifecycle.d.ts.map +1 -0
- package/dist/sandbox/providers/e2b/lifecycle.js +37 -0
- package/dist/sandbox/providers/e2b/lifecycle.js.map +1 -0
- package/dist/sandbox/providers/e2b/outcome.d.ts +11 -0
- package/dist/sandbox/providers/e2b/outcome.d.ts.map +1 -0
- package/dist/sandbox/providers/e2b/outcome.js +72 -0
- package/dist/sandbox/providers/e2b/outcome.js.map +1 -0
- package/dist/sandbox/providers/e2b/stage-files.d.ts +10 -0
- package/dist/sandbox/providers/e2b/stage-files.d.ts.map +1 -0
- package/dist/sandbox/providers/e2b/stage-files.js +45 -0
- package/dist/sandbox/providers/e2b/stage-files.js.map +1 -0
- package/dist/sandbox/providers/e2b/startup.d.ts +10 -0
- package/dist/sandbox/providers/e2b/startup.d.ts.map +1 -0
- package/dist/sandbox/providers/e2b/startup.js +26 -0
- package/dist/sandbox/providers/e2b/startup.js.map +1 -0
- package/dist/sandbox/providers/e2b/types.d.ts +9 -0
- package/dist/sandbox/providers/e2b/types.d.ts.map +1 -0
- package/dist/sandbox/providers/e2b/types.js +2 -0
- package/dist/sandbox/providers/e2b/types.js.map +1 -0
- package/dist/sandbox/providers/e2b/validation.d.ts +14 -0
- package/dist/sandbox/providers/e2b/validation.d.ts.map +1 -0
- package/dist/sandbox/providers/e2b/validation.js +60 -0
- package/dist/sandbox/providers/e2b/validation.js.map +1 -0
- package/dist/sandbox/providers/modal/auth.d.ts +2 -0
- package/dist/sandbox/providers/modal/auth.d.ts.map +1 -0
- package/dist/sandbox/providers/modal/auth.js +9 -0
- package/dist/sandbox/providers/modal/auth.js.map +1 -0
- package/dist/sandbox/providers/modal/executor.d.ts +15 -0
- package/dist/sandbox/providers/modal/executor.d.ts.map +1 -0
- package/dist/sandbox/providers/modal/executor.js +207 -0
- package/dist/sandbox/providers/modal/executor.js.map +1 -0
- package/dist/sandbox/providers/modal/live.d.ts +7 -0
- package/dist/sandbox/providers/modal/live.d.ts.map +1 -0
- package/dist/sandbox/providers/modal/live.js +50 -0
- package/dist/sandbox/providers/modal/live.js.map +1 -0
- package/dist/sandbox/providers/vercel/command.d.ts +20 -0
- package/dist/sandbox/providers/vercel/command.d.ts.map +1 -0
- package/dist/sandbox/providers/vercel/command.js +126 -0
- package/dist/sandbox/providers/vercel/command.js.map +1 -0
- package/dist/sandbox/providers/vercel/executor.d.ts +13 -0
- package/dist/sandbox/providers/vercel/executor.d.ts.map +1 -0
- package/dist/sandbox/providers/vercel/executor.js +190 -0
- package/dist/sandbox/providers/vercel/executor.js.map +1 -0
- package/dist/sandbox/providers/vercel/fetch.d.ts +6 -0
- package/dist/sandbox/providers/vercel/fetch.d.ts.map +1 -0
- package/dist/sandbox/providers/vercel/fetch.js +39 -0
- package/dist/sandbox/providers/vercel/fetch.js.map +1 -0
- package/dist/sandbox/providers/vercel/lifecycle.d.ts +13 -0
- package/dist/sandbox/providers/vercel/lifecycle.d.ts.map +1 -0
- package/dist/sandbox/providers/vercel/lifecycle.js +140 -0
- package/dist/sandbox/providers/vercel/lifecycle.js.map +1 -0
- package/dist/sandbox/providers/vercel/validation.d.ts +11 -0
- package/dist/sandbox/providers/vercel/validation.d.ts.map +1 -0
- package/dist/sandbox/providers/vercel/validation.js +30 -0
- package/dist/sandbox/providers/vercel/validation.js.map +1 -0
- package/dist/sandbox/remote-files.d.ts +11 -0
- package/dist/sandbox/remote-files.d.ts.map +1 -0
- package/dist/sandbox/remote-files.js +41 -0
- package/dist/sandbox/remote-files.js.map +1 -0
- package/dist/sandbox/request-validation.d.ts +5 -0
- package/dist/sandbox/request-validation.d.ts.map +1 -0
- package/dist/sandbox/request-validation.js +43 -0
- package/dist/sandbox/request-validation.js.map +1 -0
- package/dist/sandbox/supervisor.d.ts +42 -0
- package/dist/sandbox/supervisor.d.ts.map +1 -0
- package/dist/sandbox/supervisor.js +126 -0
- package/dist/sandbox/supervisor.js.map +1 -0
- package/dist/sandbox/timeout.d.ts +16 -0
- package/dist/sandbox/timeout.d.ts.map +1 -0
- package/dist/sandbox/timeout.js +53 -0
- package/dist/sandbox/timeout.js.map +1 -0
- package/dist/sandbox-author.bundle.js +48 -14877
- package/dist/sandbox-check.bundle.js +20155 -0
- package/dist/sandbox-verifier.bundle.js +307 -0
- package/dist/setup/e2b/build.d.ts +24 -0
- package/dist/setup/e2b/build.d.ts.map +1 -0
- package/dist/setup/e2b/build.js +46 -0
- package/dist/setup/e2b/build.js.map +1 -0
- package/dist/setup/e2b/cli.d.ts +2 -0
- package/dist/setup/e2b/cli.d.ts.map +1 -0
- package/dist/setup/e2b/cli.js +51 -0
- package/dist/setup/e2b/cli.js.map +1 -0
- package/dist/setup/e2b/index.d.ts +5 -0
- package/dist/setup/e2b/index.d.ts.map +1 -0
- package/dist/setup/e2b/index.js +4 -0
- package/dist/setup/e2b/index.js.map +1 -0
- package/dist/setup/e2b/template.d.ts +4 -0
- package/dist/setup/e2b/template.d.ts.map +1 -0
- package/dist/setup/e2b/template.js +53 -0
- package/dist/setup/e2b/template.js.map +1 -0
- package/dist/setup/vercel/cli-helpers.d.ts +22 -0
- package/dist/setup/vercel/cli-helpers.d.ts.map +1 -0
- package/dist/setup/vercel/cli-helpers.js +87 -0
- package/dist/setup/vercel/cli-helpers.js.map +1 -0
- package/dist/setup/vercel/cli-runner.d.ts +14 -0
- package/dist/setup/vercel/cli-runner.d.ts.map +1 -0
- package/dist/setup/vercel/cli-runner.js +24 -0
- package/dist/setup/vercel/cli-runner.js.map +1 -0
- package/dist/setup/vercel/cli-types.d.ts +24 -0
- package/dist/setup/vercel/cli-types.d.ts.map +1 -0
- package/dist/setup/vercel/cli-types.js +18 -0
- package/dist/setup/vercel/cli-types.js.map +1 -0
- package/dist/setup/vercel/cli.d.ts +38 -0
- package/dist/setup/vercel/cli.d.ts.map +1 -0
- package/dist/setup/vercel/cli.js +193 -0
- package/dist/setup/vercel/cli.js.map +1 -0
- package/dist/setup/vercel/index.d.ts +4 -0
- package/dist/setup/vercel/index.d.ts.map +1 -0
- package/dist/setup/vercel/index.js +3 -0
- package/dist/setup/vercel/index.js.map +1 -0
- package/dist/setup/vercel/probe.d.ts +49 -0
- package/dist/setup/vercel/probe.d.ts.map +1 -0
- package/dist/setup/vercel/probe.js +151 -0
- package/dist/setup/vercel/probe.js.map +1 -0
- package/dist/setup/vercel/profile-files.d.ts +6 -0
- package/dist/setup/vercel/profile-files.d.ts.map +1 -0
- package/dist/setup/vercel/profile-files.js +100 -0
- package/dist/setup/vercel/profile-files.js.map +1 -0
- package/dist/setup/vercel/profile-schema.d.ts +48 -0
- package/dist/setup/vercel/profile-schema.d.ts.map +1 -0
- package/dist/setup/vercel/profile-schema.js +39 -0
- package/dist/setup/vercel/profile-schema.js.map +1 -0
- package/dist/setup/vercel/profile.d.ts +18 -0
- package/dist/setup/vercel/profile.d.ts.map +1 -0
- package/dist/setup/vercel/profile.js +118 -0
- package/dist/setup/vercel/profile.js.map +1 -0
- package/dist/setup/vercel/runtime-image.d.ts +51 -0
- package/dist/setup/vercel/runtime-image.d.ts.map +1 -0
- package/dist/setup/vercel/runtime-image.js +123 -0
- package/dist/setup/vercel/runtime-image.js.map +1 -0
- package/dist/setup/vercel/selection.d.ts +18 -0
- package/dist/setup/vercel/selection.d.ts.map +1 -0
- package/dist/setup/vercel/selection.js +113 -0
- package/dist/setup/vercel/selection.js.map +1 -0
- package/dist/setup/vercel/setup-types.d.ts +34 -0
- package/dist/setup/vercel/setup-types.d.ts.map +1 -0
- package/dist/setup/vercel/setup-types.js +2 -0
- package/dist/setup/vercel/setup-types.js.map +1 -0
- package/dist/setup/vercel/setup.d.ts +17 -0
- package/dist/setup/vercel/setup.d.ts.map +1 -0
- package/dist/setup/vercel/setup.js +172 -0
- package/dist/setup/vercel/setup.js.map +1 -0
- package/dist/site/batch-activity.d.ts +5 -0
- package/dist/site/batch-activity.d.ts.map +1 -0
- package/dist/site/batch-activity.js +31 -0
- package/dist/site/batch-activity.js.map +1 -0
- package/dist/site/batch-progress.d.ts +16 -0
- package/dist/site/batch-progress.d.ts.map +1 -0
- package/dist/site/batch-progress.js +52 -0
- package/dist/site/batch-progress.js.map +1 -0
- package/dist/site/batch-routes.d.ts +30 -0
- package/dist/site/batch-routes.d.ts.map +1 -0
- package/dist/site/batch-routes.js +117 -0
- package/dist/site/batch-routes.js.map +1 -0
- package/dist/site/batch-start.d.ts +60 -0
- package/dist/site/batch-start.d.ts.map +1 -0
- package/dist/site/batch-start.js +39 -0
- package/dist/site/batch-start.js.map +1 -0
- package/dist/site/connected-repos.d.ts +25 -0
- package/dist/site/connected-repos.d.ts.map +1 -0
- package/dist/site/connected-repos.js +109 -0
- package/dist/site/connected-repos.js.map +1 -0
- package/dist/site/evaluation-start.d.ts +4 -0
- package/dist/site/evaluation-start.d.ts.map +1 -0
- package/dist/site/evaluation-start.js +19 -0
- package/dist/site/evaluation-start.js.map +1 -0
- package/dist/site/generation-config.d.ts +4 -0
- package/dist/site/generation-config.d.ts.map +1 -0
- package/dist/site/generation-config.js +16 -0
- package/dist/site/generation-config.js.map +1 -0
- package/dist/site/generation-credentials.d.ts +6 -0
- package/dist/site/generation-credentials.d.ts.map +1 -0
- package/dist/site/generation-credentials.js +101 -0
- package/dist/site/generation-credentials.js.map +1 -0
- package/dist/site/generation-run.d.ts +6 -0
- package/dist/site/generation-run.d.ts.map +1 -0
- package/dist/site/generation-run.js +22 -0
- package/dist/site/generation-run.js.map +1 -0
- package/dist/site/generation-settings.d.ts +80 -0
- package/dist/site/generation-settings.d.ts.map +1 -0
- package/dist/site/generation-settings.js +84 -0
- package/dist/site/generation-settings.js.map +1 -0
- package/dist/site/generation-subscription.d.ts +3 -0
- package/dist/site/generation-subscription.d.ts.map +1 -0
- package/dist/site/generation-subscription.js +38 -0
- package/dist/site/generation-subscription.js.map +1 -0
- package/dist/site/github-repos.d.ts +35 -0
- package/dist/site/github-repos.d.ts.map +1 -0
- package/dist/site/github-repos.js +121 -0
- package/dist/site/github-repos.js.map +1 -0
- package/dist/site/pr-candidate.d.ts +26 -0
- package/dist/site/pr-candidate.d.ts.map +1 -0
- package/dist/site/pr-candidate.js +101 -0
- package/dist/site/pr-candidate.js.map +1 -0
- package/dist/site/pr-list.d.ts +13 -0
- package/dist/site/pr-list.d.ts.map +1 -0
- package/dist/site/pr-list.js +24 -0
- package/dist/site/pr-list.js.map +1 -0
- package/dist/site/pr-routes.d.ts +26 -0
- package/dist/site/pr-routes.d.ts.map +1 -0
- package/dist/site/pr-routes.js +139 -0
- package/dist/site/pr-routes.js.map +1 -0
- package/dist/site/repo-store.d.ts +40 -0
- package/dist/site/repo-store.d.ts.map +1 -0
- package/dist/site/repo-store.js +74 -0
- package/dist/site/repo-store.js.map +1 -0
- package/dist/site/run-store.d.ts +17 -0
- package/dist/site/run-store.d.ts.map +1 -0
- package/dist/site/run-store.js +42 -0
- package/dist/site/run-store.js.map +1 -0
- package/dist/site/runtime.d.ts +27 -0
- package/dist/site/runtime.d.ts.map +1 -0
- package/dist/site/runtime.js +97 -0
- package/dist/site/runtime.js.map +1 -0
- package/dist/site/task-deletion.d.ts +7 -0
- package/dist/site/task-deletion.d.ts.map +1 -0
- package/dist/site/task-deletion.js +29 -0
- package/dist/site/task-deletion.js.map +1 -0
- package/dist/site/task-start-reservation.d.ts +4 -0
- package/dist/site/task-start-reservation.d.ts.map +1 -0
- package/dist/site/task-start-reservation.js +18 -0
- package/dist/site/task-start-reservation.js.map +1 -0
- package/dist/site/task-start.d.ts +33 -0
- package/dist/site/task-start.d.ts.map +1 -0
- package/dist/site/task-start.js +50 -0
- package/dist/site/task-start.js.map +1 -0
- package/dist/site/task-status.d.ts +36 -0
- package/dist/site/task-status.d.ts.map +1 -0
- package/dist/site/task-status.js +113 -0
- package/dist/site/task-status.js.map +1 -0
- package/dist/site/task-store.d.ts +78 -0
- package/dist/site/task-store.d.ts.map +1 -0
- package/dist/site/task-store.js +186 -0
- package/dist/site/task-store.js.map +1 -0
- package/dist/site/task-sync.d.ts +30 -0
- package/dist/site/task-sync.d.ts.map +1 -0
- package/dist/site/task-sync.js +121 -0
- package/dist/site/task-sync.js.map +1 -0
- package/dist/site/tasks.d.ts +47 -0
- package/dist/site/tasks.d.ts.map +1 -0
- package/dist/site/tasks.js +164 -0
- package/dist/site/tasks.js.map +1 -0
- package/dist/site/temporal-status.d.ts +8 -0
- package/dist/site/temporal-status.d.ts.map +1 -0
- package/dist/site/temporal-status.js +41 -0
- package/dist/site/temporal-status.js.map +1 -0
- package/dist/site/tenant.d.ts +4 -0
- package/dist/site/tenant.d.ts.map +1 -0
- package/dist/site/tenant.js +6 -0
- package/dist/site/tenant.js.map +1 -0
- package/dist/static-check.d.ts +38 -0
- package/dist/static-check.d.ts.map +1 -0
- package/dist/static-check.js +131 -0
- package/dist/static-check.js.map +1 -0
- package/dist/submission-hash.d.ts +18 -0
- package/dist/submission-hash.d.ts.map +1 -0
- package/dist/submission-hash.js +16 -0
- package/dist/submission-hash.js.map +1 -0
- package/dist/subscription-auth.d.ts +0 -7
- package/dist/subscription-auth.d.ts.map +1 -1
- package/dist/subscription-auth.js +3 -23
- package/dist/subscription-auth.js.map +1 -1
- package/dist/temporal/activities/agent-feed.d.ts +5 -0
- package/dist/temporal/activities/agent-feed.d.ts.map +1 -0
- package/dist/temporal/activities/agent-feed.js +41 -0
- package/dist/temporal/activities/agent-feed.js.map +1 -0
- package/dist/temporal/activities/agent-scripts.d.ts +16 -0
- package/dist/temporal/activities/agent-scripts.d.ts.map +1 -0
- package/dist/temporal/activities/agent-scripts.js +162 -0
- package/dist/temporal/activities/agent-scripts.js.map +1 -0
- package/dist/temporal/activities/authoring-round.d.ts +13 -0
- package/dist/temporal/activities/authoring-round.d.ts.map +1 -0
- package/dist/temporal/activities/authoring-round.js +160 -0
- package/dist/temporal/activities/authoring-round.js.map +1 -0
- package/dist/temporal/activities/constants.d.ts +5 -0
- package/dist/temporal/activities/constants.d.ts.map +1 -0
- package/dist/temporal/activities/constants.js +5 -0
- package/dist/temporal/activities/constants.js.map +1 -0
- package/dist/temporal/activities/discovery.d.ts +17 -0
- package/dist/temporal/activities/discovery.d.ts.map +1 -0
- package/dist/temporal/activities/discovery.js +153 -0
- package/dist/temporal/activities/discovery.js.map +1 -0
- package/dist/temporal/activities/drafts.d.ts +15 -0
- package/dist/temporal/activities/drafts.d.ts.map +1 -0
- package/dist/temporal/activities/drafts.js +51 -0
- package/dist/temporal/activities/drafts.js.map +1 -0
- package/dist/temporal/activities/excluded-source-prs.d.ts +9 -0
- package/dist/temporal/activities/excluded-source-prs.d.ts.map +1 -0
- package/dist/temporal/activities/excluded-source-prs.js +132 -0
- package/dist/temporal/activities/excluded-source-prs.js.map +1 -0
- package/dist/temporal/activities/export.d.ts +38 -0
- package/dist/temporal/activities/export.d.ts.map +1 -0
- package/dist/temporal/activities/export.js +79 -0
- package/dist/temporal/activities/export.js.map +1 -0
- package/dist/temporal/activities/factory.d.ts +5 -0
- package/dist/temporal/activities/factory.d.ts.map +1 -0
- package/dist/temporal/activities/factory.js +26 -0
- package/dist/temporal/activities/factory.js.map +1 -0
- package/dist/temporal/activities/gate-logs.d.ts +20 -0
- package/dist/temporal/activities/gate-logs.d.ts.map +1 -0
- package/dist/temporal/activities/gate-logs.js +62 -0
- package/dist/temporal/activities/gate-logs.js.map +1 -0
- package/dist/temporal/activities/generation-runtime.d.ts +6 -0
- package/dist/temporal/activities/generation-runtime.d.ts.map +1 -0
- package/dist/temporal/activities/generation-runtime.js +56 -0
- package/dist/temporal/activities/generation-runtime.js.map +1 -0
- package/dist/temporal/activities/harbor.d.ts +11 -0
- package/dist/temporal/activities/harbor.d.ts.map +1 -0
- package/dist/temporal/activities/harbor.js +112 -0
- package/dist/temporal/activities/harbor.js.map +1 -0
- package/dist/temporal/activities/modal-build-log.d.ts +12 -0
- package/dist/temporal/activities/modal-build-log.d.ts.map +1 -0
- package/dist/temporal/activities/modal-build-log.js +57 -0
- package/dist/temporal/activities/modal-build-log.js.map +1 -0
- package/dist/temporal/activities/prompts-authoring.d.ts +6 -0
- package/dist/temporal/activities/prompts-authoring.d.ts.map +1 -0
- package/dist/temporal/activities/prompts-authoring.js +87 -0
- package/dist/temporal/activities/prompts-authoring.js.map +1 -0
- package/dist/temporal/activities/prompts-verifier.d.ts +17 -0
- package/dist/temporal/activities/prompts-verifier.d.ts.map +1 -0
- package/dist/temporal/activities/prompts-verifier.js +66 -0
- package/dist/temporal/activities/prompts-verifier.js.map +1 -0
- package/dist/temporal/activities/provenance.d.ts +4 -0
- package/dist/temporal/activities/provenance.d.ts.map +1 -0
- package/dist/temporal/activities/provenance.js +18 -0
- package/dist/temporal/activities/provenance.js.map +1 -0
- package/dist/temporal/activities/replay.d.ts +20 -0
- package/dist/temporal/activities/replay.d.ts.map +1 -0
- package/dist/temporal/activities/replay.js +135 -0
- package/dist/temporal/activities/replay.js.map +1 -0
- package/dist/temporal/activities/round-outcome.d.ts +63 -0
- package/dist/temporal/activities/round-outcome.d.ts.map +1 -0
- package/dist/temporal/activities/round-outcome.js +127 -0
- package/dist/temporal/activities/round-outcome.js.map +1 -0
- package/dist/temporal/activities/runtime.d.ts +38 -0
- package/dist/temporal/activities/runtime.d.ts.map +1 -0
- package/dist/temporal/activities/runtime.js +146 -0
- package/dist/temporal/activities/runtime.js.map +1 -0
- package/dist/temporal/activities/session-verify.d.ts +45 -0
- package/dist/temporal/activities/session-verify.d.ts.map +1 -0
- package/dist/temporal/activities/session-verify.js +91 -0
- package/dist/temporal/activities/session-verify.js.map +1 -0
- package/dist/temporal/activities/submissions.d.ts +9 -0
- package/dist/temporal/activities/submissions.d.ts.map +1 -0
- package/dist/temporal/activities/submissions.js +28 -0
- package/dist/temporal/activities/submissions.js.map +1 -0
- package/dist/temporal/activities/task-compiler.d.ts +21 -0
- package/dist/temporal/activities/task-compiler.d.ts.map +1 -0
- package/dist/temporal/activities/task-compiler.js +63 -0
- package/dist/temporal/activities/task-compiler.js.map +1 -0
- package/dist/temporal/activities/types.d.ts +57 -0
- package/dist/temporal/activities/types.d.ts.map +1 -0
- package/dist/temporal/activities/types.js +2 -0
- package/dist/temporal/activities/types.js.map +1 -0
- package/dist/temporal/activities/verifier-material.d.ts +18 -0
- package/dist/temporal/activities/verifier-material.d.ts.map +1 -0
- package/dist/temporal/activities/verifier-material.js +49 -0
- package/dist/temporal/activities/verifier-material.js.map +1 -0
- package/dist/temporal/activities/verifier-outcome.d.ts +19 -0
- package/dist/temporal/activities/verifier-outcome.d.ts.map +1 -0
- package/dist/temporal/activities/verifier-outcome.js +70 -0
- package/dist/temporal/activities/verifier-outcome.js.map +1 -0
- package/dist/temporal/activities/verifier-round.d.ts +8 -0
- package/dist/temporal/activities/verifier-round.d.ts.map +1 -0
- package/dist/temporal/activities/verifier-round.js +110 -0
- package/dist/temporal/activities/verifier-round.js.map +1 -0
- package/dist/temporal/activities/verify-harbor.d.ts +15 -0
- package/dist/temporal/activities/verify-harbor.d.ts.map +1 -0
- package/dist/temporal/activities/verify-harbor.js +193 -0
- package/dist/temporal/activities/verify-harbor.js.map +1 -0
- package/dist/temporal/activities/verify.d.ts +11 -0
- package/dist/temporal/activities/verify.d.ts.map +1 -0
- package/dist/temporal/activities/verify.js +151 -0
- package/dist/temporal/activities/verify.js.map +1 -0
- package/dist/temporal/activities.d.ts +3 -39
- package/dist/temporal/activities.d.ts.map +1 -1
- package/dist/temporal/activities.js +2 -839
- package/dist/temporal/activities.js.map +1 -1
- package/dist/temporal/connection.d.ts.map +1 -1
- package/dist/temporal/connection.js +3 -2
- package/dist/temporal/connection.js.map +1 -1
- package/dist/temporal/worker-main.js +19 -3
- package/dist/temporal/worker-main.js.map +1 -1
- package/dist/temporal/workflow/activity-proxies.d.ts +3 -0
- package/dist/temporal/workflow/activity-proxies.d.ts.map +1 -0
- package/dist/temporal/workflow/activity-proxies.js +45 -0
- package/dist/temporal/workflow/activity-proxies.js.map +1 -0
- package/dist/temporal/workflow/authoring-stage.d.ts +5 -0
- package/dist/temporal/workflow/authoring-stage.d.ts.map +1 -0
- package/dist/temporal/workflow/authoring-stage.js +64 -0
- package/dist/temporal/workflow/authoring-stage.js.map +1 -0
- package/dist/temporal/workflow/candidate-tracker.d.ts +28 -0
- package/dist/temporal/workflow/candidate-tracker.d.ts.map +1 -0
- package/dist/temporal/workflow/candidate-tracker.js +85 -0
- package/dist/temporal/workflow/candidate-tracker.js.map +1 -0
- package/dist/temporal/workflow/candidate.d.ts +6 -0
- package/dist/temporal/workflow/candidate.d.ts.map +1 -0
- package/dist/temporal/workflow/candidate.js +54 -0
- package/dist/temporal/workflow/candidate.js.map +1 -0
- package/dist/temporal/workflow/discovery.d.ts +19 -0
- package/dist/temporal/workflow/discovery.d.ts.map +1 -0
- package/dist/temporal/workflow/discovery.js +137 -0
- package/dist/temporal/workflow/discovery.js.map +1 -0
- package/dist/temporal/workflow/failures.d.ts +5 -0
- package/dist/temporal/workflow/failures.d.ts.map +1 -0
- package/dist/temporal/workflow/failures.js +28 -0
- package/dist/temporal/workflow/failures.js.map +1 -0
- package/dist/temporal/workflow/run.d.ts +5 -0
- package/dist/temporal/workflow/run.d.ts.map +1 -0
- package/dist/temporal/workflow/run.js +154 -0
- package/dist/temporal/workflow/run.js.map +1 -0
- package/dist/temporal/workflow/stage.d.ts +26 -0
- package/dist/temporal/workflow/stage.d.ts.map +1 -0
- package/dist/temporal/workflow/stage.js +16 -0
- package/dist/temporal/workflow/stage.js.map +1 -0
- package/dist/temporal/workflow/verification-stage.d.ts +11 -0
- package/dist/temporal/workflow/verification-stage.d.ts.map +1 -0
- package/dist/temporal/workflow/verification-stage.js +20 -0
- package/dist/temporal/workflow/verification-stage.js.map +1 -0
- package/dist/temporal/workflow.d.ts +13 -5
- package/dist/temporal/workflow.d.ts.map +1 -1
- package/dist/temporal/workflow.js +38 -420
- package/dist/temporal/workflow.js.map +1 -1
- package/dist/verifier-fix.d.ts +20 -0
- package/dist/verifier-fix.d.ts.map +1 -0
- package/dist/verifier-fix.js +62 -0
- package/dist/verifier-fix.js.map +1 -0
- package/dist/verify-report.d.ts +9 -0
- package/dist/verify-report.d.ts.map +1 -0
- package/dist/verify-report.js +135 -0
- package/dist/verify-report.js.map +1 -0
- package/dist/viewer/archived-keys.d.ts +9 -0
- package/dist/viewer/archived-keys.d.ts.map +1 -0
- package/dist/viewer/archived-keys.js +36 -0
- package/dist/viewer/archived-keys.js.map +1 -0
- package/dist/viewer/archived.d.ts +20 -0
- package/dist/viewer/archived.d.ts.map +1 -0
- package/dist/viewer/archived.js +238 -0
- package/dist/viewer/archived.js.map +1 -0
- package/dist/viewer/artifacts.d.ts +8 -0
- package/dist/viewer/artifacts.d.ts.map +1 -0
- package/dist/viewer/artifacts.js +96 -0
- package/dist/viewer/artifacts.js.map +1 -0
- package/dist/viewer/bundle.d.ts +9 -0
- package/dist/viewer/bundle.d.ts.map +1 -0
- package/dist/viewer/bundle.js +79 -0
- package/dist/viewer/bundle.js.map +1 -0
- package/dist/viewer/candidates.d.ts +9 -0
- package/dist/viewer/candidates.d.ts.map +1 -0
- package/dist/viewer/candidates.js +137 -0
- package/dist/viewer/candidates.js.map +1 -0
- package/dist/viewer/local-server.d.ts +11 -0
- package/dist/viewer/local-server.d.ts.map +1 -0
- package/dist/viewer/local-server.js +64 -0
- package/dist/viewer/local-server.js.map +1 -0
- package/dist/viewer/task-files.d.ts +10 -0
- package/dist/viewer/task-files.d.ts.map +1 -0
- package/dist/viewer/task-files.js +120 -0
- package/dist/viewer/task-files.js.map +1 -0
- package/dist/viewer/types.d.ts +78 -0
- package/dist/viewer/types.d.ts.map +1 -0
- package/dist/viewer/types.js +36 -0
- package/dist/viewer/types.js.map +1 -0
- package/docs/api.md +168 -0
- package/docs/evaluations.md +42 -15
- package/docs/operations.md +292 -30
- package/docs/solver-evaluations.md +63 -0
- package/docs/task-construction.md +24 -12
- package/drizzle/0000_glossy_johnny_blaze.sql +97 -0
- package/drizzle/0001_evaluation_records.sql +5 -0
- package/drizzle/0002_task_deletion.sql +1 -0
- package/drizzle/0003_drop_repo_runs.sql +3 -0
- package/drizzle/0004_preserve_batch_runs.sql +8 -0
- package/drizzle/0005_api_keys.sql +15 -0
- package/drizzle/meta/0000_snapshot.json +776 -0
- package/drizzle/meta/0001_snapshot.json +807 -0
- package/drizzle/meta/0002_snapshot.json +813 -0
- package/drizzle/meta/0003_snapshot.json +726 -0
- package/drizzle/meta/0004_snapshot.json +813 -0
- package/drizzle/meta/0005_snapshot.json +919 -0
- package/drizzle/meta/_journal.json +48 -0
- package/package.json +43 -29
- package/src/extensions/authoring.ts +83 -72
- package/src/extensions/discovery.ts +14 -5
- package/src/extensions/shared/deliverable.ts +153 -0
- package/src/extensions/shared/mailbox.ts +128 -0
- package/src/extensions/shared/schemas.ts +101 -0
- package/src/extensions/shared/static-check.ts +103 -0
- package/src/extensions/verifier.ts +88 -0
- package/src/skills/selfbench/SKILL.md +71 -13
- package/biome.json +0 -18
- package/dist/agent-smoke-main.d.ts +0 -3
- package/dist/agent-smoke-main.d.ts.map +0 -1
- package/dist/agent-smoke-main.js +0 -55
- package/dist/agent-smoke-main.js.map +0 -1
- package/dist/agent-smoke.d.ts +0 -18
- package/dist/agent-smoke.d.ts.map +0 -1
- package/dist/agent-smoke.js +0 -103
- package/dist/agent-smoke.js.map +0 -1
- package/dist/codex-review.d.ts +0 -41
- package/dist/codex-review.d.ts.map +0 -1
- package/dist/codex-review.js +0 -193
- package/dist/codex-review.js.map +0 -1
- package/dist/docker-executor.d.ts +0 -11
- package/dist/docker-executor.d.ts.map +0 -1
- package/dist/docker-executor.js +0 -103
- package/dist/docker-executor.js.map +0 -1
- package/dist/eval-main.d.ts +0 -3
- package/dist/eval-main.d.ts.map +0 -1
- package/dist/eval-main.js +0 -81
- package/dist/eval-main.js.map +0 -1
- package/dist/evaluate.d.ts +0 -25
- package/dist/evaluate.d.ts.map +0 -1
- package/dist/evaluate.js +0 -214
- package/dist/evaluate.js.map +0 -1
- package/dist/modal-auth.d.ts +0 -2
- package/dist/modal-auth.d.ts.map +0 -1
- package/dist/modal-auth.js +0 -9
- package/dist/modal-auth.js.map +0 -1
- package/dist/modal-executor.d.ts +0 -12
- package/dist/modal-executor.d.ts.map +0 -1
- package/dist/modal-executor.js +0 -152
- package/dist/modal-executor.js.map +0 -1
- package/dist/reaudit-main.d.ts +0 -3
- package/dist/reaudit-main.d.ts.map +0 -1
- package/dist/reaudit-main.js +0 -158
- package/dist/reaudit-main.js.map +0 -1
- package/dist/repair-main.d.ts +0 -3
- package/dist/repair-main.d.ts.map +0 -1
- package/dist/repair-main.js +0 -171
- package/dist/repair-main.js.map +0 -1
- package/dist/review/assets/index-CZ3G759O.js +0 -1568
- package/dist/review/assets/index-DgSLzX8l.css +0 -1
- package/dist/sandbox-author.d.ts +0 -3
- package/dist/sandbox-author.d.ts.map +0 -1
- package/dist/sandbox-author.js +0 -17
- package/dist/sandbox-author.js.map +0 -1
- package/dist/sandbox-repair.bundle.js +0 -315
- package/dist/sandbox-repair.d.ts +0 -3
- package/dist/sandbox-repair.d.ts.map +0 -1
- package/dist/sandbox-repair.js +0 -121
- package/dist/sandbox-repair.js.map +0 -1
- package/dist/sandbox-review.bundle.js +0 -14462
- package/dist/sandbox-review.d.ts +0 -2
- package/dist/sandbox-review.d.ts.map +0 -1
- package/dist/sandbox-review.js +0 -18
- package/dist/sandbox-review.js.map +0 -1
- package/dist/sandbox-timeout.d.ts +0 -11
- package/dist/sandbox-timeout.d.ts.map +0 -1
- package/dist/sandbox-timeout.js +0 -41
- package/dist/sandbox-timeout.js.map +0 -1
- package/dist/sandbox-validation-repair.bundle.js +0 -14778
- package/dist/sandbox-validation-repair.d.ts +0 -3
- package/dist/sandbox-validation-repair.d.ts.map +0 -1
- package/dist/sandbox-validation-repair.js +0 -136
- package/dist/sandbox-validation-repair.js.map +0 -1
- package/dist/sandbox.d.ts +0 -39
- package/dist/sandbox.d.ts.map +0 -1
- package/dist/sandbox.js +0 -15
- package/dist/sandbox.js.map +0 -1
- package/dist/validate-main.d.ts +0 -3
- package/dist/validate-main.d.ts.map +0 -1
- package/dist/validate-main.js +0 -143
- package/dist/validate-main.js.map +0 -1
- package/dist/validation-repair.d.ts +0 -10
- package/dist/validation-repair.d.ts.map +0 -1
- package/dist/validation-repair.js +0 -72
- package/dist/validation-repair.js.map +0 -1
- package/dist/vercel-cli.d.ts +0 -64
- package/dist/vercel-cli.d.ts.map +0 -1
- package/dist/vercel-cli.js +0 -303
- package/dist/vercel-cli.js.map +0 -1
- package/dist/vercel-command.d.ts +0 -14
- package/dist/vercel-command.d.ts.map +0 -1
- package/dist/vercel-command.js +0 -93
- package/dist/vercel-command.js.map +0 -1
- package/dist/vercel-executor.d.ts +0 -14
- package/dist/vercel-executor.d.ts.map +0 -1
- package/dist/vercel-executor.js +0 -356
- package/dist/vercel-executor.js.map +0 -1
- package/dist/vercel-fetch.d.ts +0 -6
- package/dist/vercel-fetch.d.ts.map +0 -1
- package/dist/vercel-fetch.js +0 -39
- package/dist/vercel-fetch.js.map +0 -1
- package/dist/vercel-profile.d.ts +0 -34
- package/dist/vercel-profile.d.ts.map +0 -1
- package/dist/vercel-profile.js +0 -244
- package/dist/vercel-profile.js.map +0 -1
- package/dist/vercel-runtime-image.d.ts +0 -51
- package/dist/vercel-runtime-image.d.ts.map +0 -1
- package/dist/vercel-runtime-image.js +0 -123
- package/dist/vercel-runtime-image.js.map +0 -1
- package/dist/vercel-setup-probe.d.ts +0 -49
- package/dist/vercel-setup-probe.d.ts.map +0 -1
- package/dist/vercel-setup-probe.js +0 -151
- package/dist/vercel-setup-probe.js.map +0 -1
- package/dist/vercel-setup.d.ts +0 -42
- package/dist/vercel-setup.d.ts.map +0 -1
- package/dist/vercel-setup.js +0 -280
- package/dist/vercel-setup.js.map +0 -1
- package/review/index.html +0 -14
- package/review/src/App.tsx +0 -338
- package/review/src/export-loader.test.ts +0 -57
- package/review/src/export-loader.ts +0 -151
- package/review/src/main.tsx +0 -15
- package/review/src/styles.css +0 -162
- package/review/src/types.ts +0 -47
- package/review/tsconfig.json +0 -16
- package/review/vite.config.ts +0 -20
- package/scripts/verify-package.ts +0 -81
- package/scripts/write-build-metadata.ts +0 -27
- package/src/agent-smoke-main.ts +0 -64
- package/src/agent-smoke.ts +0 -138
- package/src/api-main.ts +0 -12
- package/src/api.ts +0 -290
- package/src/artifacts.ts +0 -357
- package/src/audit.ts +0 -106
- package/src/build-metadata.ts +0 -3
- package/src/cli.ts +0 -420
- package/src/codex-review.ts +0 -229
- package/src/config.ts +0 -239
- package/src/contracts.ts +0 -231
- package/src/coupling.ts +0 -259
- package/src/docker-executor.ts +0 -115
- package/src/eval-main.ts +0 -93
- package/src/evaluate.ts +0 -300
- package/src/github.ts +0 -26
- package/src/harbor-environment.ts +0 -17
- package/src/harbor-results.ts +0 -144
- package/src/harbor-task.ts +0 -528
- package/src/hash.ts +0 -5
- package/src/modal-auth.ts +0 -11
- package/src/modal-executor.ts +0 -176
- package/src/parallel.ts +0 -24
- package/src/process.ts +0 -167
- package/src/provenance.ts +0 -457
- package/src/providers.ts +0 -22
- package/src/reaudit-main.ts +0 -198
- package/src/repair-main.ts +0 -209
- package/src/repair.ts +0 -55
- package/src/run-wait.ts +0 -40
- package/src/sandbox-author.ts +0 -19
- package/src/sandbox-repair.ts +0 -156
- package/src/sandbox-review.ts +0 -19
- package/src/sandbox-timeout.ts +0 -56
- package/src/sandbox-validation-repair.ts +0 -177
- package/src/sandbox.ts +0 -61
- package/src/subscription-auth.ts +0 -80
- package/src/temporal/activities.ts +0 -1247
- package/src/temporal/connection.ts +0 -23
- package/src/temporal/worker-main.ts +0 -28
- package/src/temporal/workflow.ts +0 -519
- package/src/terminal-prompts.ts +0 -197
- package/src/terminal-reporter.ts +0 -126
- package/src/validate-main.ts +0 -177
- package/src/validation-repair.ts +0 -94
- package/src/vercel-cli.ts +0 -403
- package/src/vercel-command.ts +0 -127
- package/src/vercel-executor.ts +0 -448
- package/src/vercel-fetch.ts +0 -45
- package/src/vercel-profile.ts +0 -319
- package/src/vercel-runtime-image.ts +0 -204
- package/src/vercel-setup-probe.ts +0 -229
- package/src/vercel-setup.ts +0 -419
- package/tsconfig.build.json +0 -13
- package/tsconfig.json +0 -21
|
@@ -0,0 +1,91 @@
|
|
|
1
|
+
import { CancelledFailure, Context } from "@temporalio/activity";
|
|
2
|
+
import { taskDefinitionSchema, } from "../../contracts.js";
|
|
3
|
+
import { superviseMailbox, } from "../../sandbox/supervisor.js";
|
|
4
|
+
import { matchingGreenVerify, submissionHash } from "../../submission-hash.js";
|
|
5
|
+
import { assertVerifierFix } from "../../verifier-fix.js";
|
|
6
|
+
import { renderVerifyReport, verifyReportSummary } from "../../verify-report.js";
|
|
7
|
+
import { materializeDraft } from "./drafts.js";
|
|
8
|
+
import { compileAndVerify } from "./verify.js";
|
|
9
|
+
/**
|
|
10
|
+
* Worker side of the agent's `verify` tool: each mailbox request becomes a draft, goes through the
|
|
11
|
+
* same compile/audit/build/smoke/nop/oracle as a submission, and is archived per call. A later
|
|
12
|
+
* submission with the same payload hash reuses the green result.
|
|
13
|
+
*/
|
|
14
|
+
export class SessionVerifier {
|
|
15
|
+
records = [];
|
|
16
|
+
#context;
|
|
17
|
+
constructor(context) {
|
|
18
|
+
this.#context = context;
|
|
19
|
+
}
|
|
20
|
+
/** Supervises the live sandbox mailbox for the duration of the agent command. */
|
|
21
|
+
supervise(sandbox, exited) {
|
|
22
|
+
return superviseMailbox(sandbox, exited, {
|
|
23
|
+
handle: (request) => this.handle(request),
|
|
24
|
+
isFatal: (error) => error instanceof CancelledFailure,
|
|
25
|
+
onPoll: () => Context.current().heartbeat(`mailbox ${this.records.length} verifies`),
|
|
26
|
+
}).then(() => undefined);
|
|
27
|
+
}
|
|
28
|
+
async handle(request) {
|
|
29
|
+
const { store, run, candidate, stage, round } = this.#context;
|
|
30
|
+
const index = this.records.length + 1;
|
|
31
|
+
const prefix = `${this.#context.prefix}/verify-${index}`;
|
|
32
|
+
const goldPatch = this.#context.original?.goldPatch ?? request.goldPatch;
|
|
33
|
+
if (goldPatch === undefined) {
|
|
34
|
+
return { id: request.id, kind: "error", message: "verify request has no gold patch" };
|
|
35
|
+
}
|
|
36
|
+
const definitionJson = `${JSON.stringify(request.definition, null, 2)}\n`;
|
|
37
|
+
if (this.#context.original) {
|
|
38
|
+
const boundary = fixBoundaryError(this.#context.original, request, goldPatch);
|
|
39
|
+
if (boundary) {
|
|
40
|
+
return { id: request.id, kind: "error", message: boundary };
|
|
41
|
+
}
|
|
42
|
+
}
|
|
43
|
+
await store.put(`${prefix}/request.json`, Buffer.from(`${JSON.stringify(request)}\n`), "application/json");
|
|
44
|
+
const draft = await materializeDraft(store, prefix, candidate.candidateId, definitionJson, request.testPatch, goldPatch);
|
|
45
|
+
const outcome = await compileAndVerify(store, this.#context.harborEnvironment, { run, candidate, task: draft, stage, round }, prefix);
|
|
46
|
+
this.records.push({
|
|
47
|
+
index,
|
|
48
|
+
hash: submissionHash({
|
|
49
|
+
definition: request.definition,
|
|
50
|
+
testPatch: request.testPatch,
|
|
51
|
+
goldPatch,
|
|
52
|
+
}),
|
|
53
|
+
green: outcome.report.green,
|
|
54
|
+
report: outcome.reportRef,
|
|
55
|
+
...(outcome.task ? { task: outcome.task } : {}),
|
|
56
|
+
});
|
|
57
|
+
return {
|
|
58
|
+
id: request.id,
|
|
59
|
+
kind: "report",
|
|
60
|
+
green: outcome.report.green,
|
|
61
|
+
summary: verifyReportSummary(outcome.report),
|
|
62
|
+
rendered: renderVerifyReport(outcome.report),
|
|
63
|
+
};
|
|
64
|
+
}
|
|
65
|
+
/** The green verify matching a submission, if the agent submitted exactly what it verified. */
|
|
66
|
+
verified(definition, testPatch, goldPatch) {
|
|
67
|
+
const match = matchingGreenVerify(submissionHash({ definition, testPatch, goldPatch }), this.records);
|
|
68
|
+
return match?.task ? { report: match.report, task: match.task } : undefined;
|
|
69
|
+
}
|
|
70
|
+
}
|
|
71
|
+
function fixBoundaryError(original, request, goldPatch) {
|
|
72
|
+
const parsed = taskDefinitionSchema.safeParse(request.definition);
|
|
73
|
+
if (!parsed.success) {
|
|
74
|
+
return `fixed definition is invalid: ${parsed.error.issues.map((issue) => issue.message).join("; ")}`;
|
|
75
|
+
}
|
|
76
|
+
try {
|
|
77
|
+
assertVerifierFix({
|
|
78
|
+
original: original.definition,
|
|
79
|
+
fixed: parsed.data,
|
|
80
|
+
originalTestPatch: original.testPatch,
|
|
81
|
+
fixedTestPatch: request.testPatch,
|
|
82
|
+
originalGoldPatch: original.goldPatch,
|
|
83
|
+
fixedGoldPatch: goldPatch,
|
|
84
|
+
});
|
|
85
|
+
return undefined;
|
|
86
|
+
}
|
|
87
|
+
catch (error) {
|
|
88
|
+
return error instanceof Error ? error.message : String(error);
|
|
89
|
+
}
|
|
90
|
+
}
|
|
91
|
+
//# sourceMappingURL=session-verify.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"session-verify.js","sourceRoot":"","sources":["../../../src/temporal/activities/session-verify.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,gBAAgB,EAAE,OAAO,EAAE,MAAM,sBAAsB,CAAC;AAGjE,OAAO,EAKL,oBAAoB,GAErB,MAAM,oBAAoB,CAAC;AAE5B,OAAO,EAGL,gBAAgB,GACjB,MAAM,6BAA6B,CAAC;AACrC,OAAO,EAAE,mBAAmB,EAAE,cAAc,EAAE,MAAM,0BAA0B,CAAC;AAC/E,OAAO,EAAE,iBAAiB,EAAE,MAAM,uBAAuB,CAAC;AAC1D,OAAO,EAAE,kBAAkB,EAAE,mBAAmB,EAAE,MAAM,wBAAwB,CAAC;AACjF,OAAO,EAAE,gBAAgB,EAAqB,MAAM,aAAa,CAAC;AAClE,OAAO,EAAE,gBAAgB,EAAE,MAAM,aAAa,CAAC;AA4B/C;;;;GAIG;AACH,MAAM,OAAO,eAAe;IACjB,OAAO,GAA0B,EAAE,CAAC;IACpC,QAAQ,CAAuB;IAExC,YAAY,OAA6B;QACvC,IAAI,CAAC,QAAQ,GAAG,OAAO,CAAC;IAC1B,CAAC;IAED,iFAAiF;IACjF,SAAS,CAAC,OAAoB,EAAE,MAAmB;QACjD,OAAO,gBAAgB,CAAC,OAAO,EAAE,MAAM,EAAE;YACvC,MAAM,EAAE,CAAC,OAAO,EAAE,EAAE,CAAC,IAAI,CAAC,MAAM,CAAC,OAAO,CAAC;YACzC,OAAO,EAAE,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,YAAY,gBAAgB;YACrD,MAAM,EAAE,GAAG,EAAE,CAAC,OAAO,CAAC,OAAO,EAAE,CAAC,SAAS,CAAC,WAAW,IAAI,CAAC,OAAO,CAAC,MAAM,WAAW,CAAC;SACrF,CAAC,CAAC,IAAI,CAAC,GAAG,EAAE,CAAC,SAAS,CAAC,CAAC;IAC3B,CAAC;IAED,KAAK,CAAC,MAAM,CAAC,OAAuB;QAClC,MAAM,EAAE,KAAK,EAAE,GAAG,EAAE,SAAS,EAAE,KAAK,EAAE,KAAK,EAAE,GAAG,IAAI,CAAC,QAAQ,CAAC;QAC9D,MAAM,KAAK,GAAG,IAAI,CAAC,OAAO,CAAC,MAAM,GAAG,CAAC,CAAC;QACtC,MAAM,MAAM,GAAG,GAAG,IAAI,CAAC,QAAQ,CAAC,MAAM,WAAW,KAAK,EAAE,CAAC;QACzD,MAAM,SAAS,GAAG,IAAI,CAAC,QAAQ,CAAC,QAAQ,EAAE,SAAS,IAAI,OAAO,CAAC,SAAS,CAAC;QACzE,IAAI,SAAS,KAAK,SAAS,EAAE,CAAC;YAC5B,OAAO,EAAE,EAAE,EAAE,OAAO,CAAC,EAAE,EAAE,IAAI,EAAE,OAAO,EAAE,OAAO,EAAE,kCAAkC,EAAE,CAAC;QACxF,CAAC;QACD,MAAM,cAAc,GAAG,GAAG,IAAI,CAAC,SAAS,CAAC,OAAO,CAAC,UAAU,EAAE,IAAI,EAAE,CAAC,CAAC,IAAI,CAAC;QAC1E,IAAI,IAAI,CAAC,QAAQ,CAAC,QAAQ,EAAE,CAAC;YAC3B,MAAM,QAAQ,GAAG,gBAAgB,CAAC,IAAI,CAAC,QAAQ,CAAC,QAAQ,EAAE,OAAO,EAAE,SAAS,CAAC,CAAC;YAC9E,IAAI,QAAQ,EAAE,CAAC;gBACb,OAAO,EAAE,EAAE,EAAE,OAAO,CAAC,EAAE,EAAE,IAAI,EAAE,OAAO,EAAE,OAAO,EAAE,QAAQ,EAAE,CAAC;YAC9D,CAAC;QACH,CAAC;QACD,MAAM,KAAK,CAAC,GAAG,CACb,GAAG,MAAM,eAAe,EACxB,MAAM,CAAC,IAAI,CAAC,GAAG,IAAI,CAAC,SAAS,CAAC,OAAO,CAAC,IAAI,CAAC,EAC3C,kBAAkB,CACnB,CAAC;QACF,MAAM,KAAK,GAAG,MAAM,gBAAgB,CAClC,KAAK,EACL,MAAM,EACN,SAAS,CAAC,WAAW,EACrB,cAAc,EACd,OAAO,CAAC,SAAS,EACjB,SAAS,CACV,CAAC;QACF,MAAM,OAAO,GAAG,MAAM,gBAAgB,CACpC,KAAK,EACL,IAAI,CAAC,QAAQ,CAAC,iBAAiB,EAC/B,EAAE,GAAG,EAAE,SAAS,EAAE,IAAI,EAAE,KAAK,EAAE,KAAK,EAAE,KAAK,EAAE,EAC7C,MAAM,CACP,CAAC;QACF,IAAI,CAAC,OAAO,CAAC,IAAI,CAAC;YAChB,KAAK;YACL,IAAI,EAAE,cAAc,CAAC;gBACnB,UAAU,EAAE,OAAO,CAAC,UAAU;gBAC9B,SAAS,EAAE,OAAO,CAAC,SAAS;gBAC5B,SAAS;aACV,CAAC;YACF,KAAK,EAAE,OAAO,CAAC,MAAM,CAAC,KAAK;YAC3B,MAAM,EAAE,OAAO,CAAC,SAAS;YACzB,GAAG,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,IAAI,EAAE,OAAO,CAAC,IAAI,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;SAChD,CAAC,CAAC;QACH,OAAO;YACL,EAAE,EAAE,OAAO,CAAC,EAAE;YACd,IAAI,EAAE,QAAQ;YACd,KAAK,EAAE,OAAO,CAAC,MAAM,CAAC,KAAK;YAC3B,OAAO,EAAE,mBAAmB,CAAC,OAAO,CAAC,MAAM,CAAC;YAC5C,QAAQ,EAAE,kBAAkB,CAAC,OAAO,CAAC,MAAM,CAAC;SAC7C,CAAC;IACJ,CAAC;IAED,+FAA+F;IAC/F,QAAQ,CACN,UAAmB,EACnB,SAAiB,EACjB,SAAiB;QAEjB,MAAM,KAAK,GAAG,mBAAmB,CAC/B,cAAc,CAAC,EAAE,UAAU,EAAE,SAAS,EAAE,SAAS,EAAE,CAAC,EACpD,IAAI,CAAC,OAAO,CACb,CAAC;QACF,OAAO,KAAK,EAAE,IAAI,CAAC,CAAC,CAAC,EAAE,MAAM,EAAE,KAAK,CAAC,MAAM,EAAE,IAAI,EAAE,KAAK,CAAC,IAAI,EAAE,CAAC,CAAC,CAAC,SAAS,CAAC;IAC9E,CAAC;CACF;AAED,SAAS,gBAAgB,CACvB,QAAsB,EACtB,OAAuB,EACvB,SAAiB;IAEjB,MAAM,MAAM,GAAG,oBAAoB,CAAC,SAAS,CAAC,OAAO,CAAC,UAAU,CAAC,CAAC;IAClE,IAAI,CAAC,MAAM,CAAC,OAAO,EAAE,CAAC;QACpB,OAAO,gCAAgC,MAAM,CAAC,KAAK,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC;IACxG,CAAC;IACD,IAAI,CAAC;QACH,iBAAiB,CAAC;YAChB,QAAQ,EAAE,QAAQ,CAAC,UAAU;YAC7B,KAAK,EAAE,MAAM,CAAC,IAAI;YAClB,iBAAiB,EAAE,QAAQ,CAAC,SAAS;YACrC,cAAc,EAAE,OAAO,CAAC,SAAS;YACjC,iBAAiB,EAAE,QAAQ,CAAC,SAAS;YACrC,cAAc,EAAE,SAAS;SAC1B,CAAC,CAAC;QACH,OAAO,SAAS,CAAC;IACnB,CAAC;IAAC,OAAO,KAAK,EAAE,CAAC;QACf,OAAO,KAAK,YAAY,KAAK,CAAC,CAAC,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC,CAAC;IAChE,CAAC;AACH,CAAC"}
|
|
@@ -0,0 +1,9 @@
|
|
|
1
|
+
export interface SubmissionFiles {
|
|
2
|
+
readonly taskId: string | undefined;
|
|
3
|
+
readonly definition: unknown;
|
|
4
|
+
readonly testPatch: string;
|
|
5
|
+
readonly goldPatch: string;
|
|
6
|
+
}
|
|
7
|
+
/** Parses a sandbox submission (definition.json bytes + source bundle); undefined when unreadable. */
|
|
8
|
+
export declare function readSubmission(definitionBytes: Uint8Array, sourceBundle: Uint8Array): Promise<SubmissionFiles | undefined>;
|
|
9
|
+
//# sourceMappingURL=submissions.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"submissions.d.ts","sourceRoot":"","sources":["../../../src/temporal/activities/submissions.ts"],"names":[],"mappings":"AAKA,MAAM,WAAW,eAAe;IAC9B,QAAQ,CAAC,MAAM,EAAE,MAAM,GAAG,SAAS,CAAC;IACpC,QAAQ,CAAC,UAAU,EAAE,OAAO,CAAC;IAC7B,QAAQ,CAAC,SAAS,EAAE,MAAM,CAAC;IAC3B,QAAQ,CAAC,SAAS,EAAE,MAAM,CAAC;CAC5B;AAED,sGAAsG;AACtG,wBAAsB,cAAc,CAClC,eAAe,EAAE,UAAU,EAC3B,YAAY,EAAE,UAAU,GACvB,OAAO,CAAC,eAAe,GAAG,SAAS,CAAC,CAoBtC"}
|
|
@@ -0,0 +1,28 @@
|
|
|
1
|
+
import { mkdir, writeFile } from "node:fs/promises";
|
|
2
|
+
import { join } from "node:path";
|
|
3
|
+
import { extractRegularArchive } from "../../archive.js";
|
|
4
|
+
import { readTaskPatches, withTemporaryDirectory } from "./runtime.js";
|
|
5
|
+
/** Parses a sandbox submission (definition.json bytes + source bundle); undefined when unreadable. */
|
|
6
|
+
export async function readSubmission(definitionBytes, sourceBundle) {
|
|
7
|
+
try {
|
|
8
|
+
const definition = JSON.parse(Buffer.from(definitionBytes).toString("utf8"));
|
|
9
|
+
const patches = await withTemporaryDirectory("selfbench-submission-", async (root) => {
|
|
10
|
+
const archive = join(root, "source-task.tar.gz");
|
|
11
|
+
const authored = join(root, "authored");
|
|
12
|
+
await mkdir(authored);
|
|
13
|
+
await writeFile(archive, sourceBundle);
|
|
14
|
+
await extractRegularArchive(archive, authored);
|
|
15
|
+
return await readTaskPatches(authored);
|
|
16
|
+
});
|
|
17
|
+
const taskId = definition?.taskId;
|
|
18
|
+
return {
|
|
19
|
+
taskId: typeof taskId === "string" && taskId ? taskId : undefined,
|
|
20
|
+
definition,
|
|
21
|
+
...patches,
|
|
22
|
+
};
|
|
23
|
+
}
|
|
24
|
+
catch {
|
|
25
|
+
return undefined;
|
|
26
|
+
}
|
|
27
|
+
}
|
|
28
|
+
//# sourceMappingURL=submissions.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"submissions.js","sourceRoot":"","sources":["../../../src/temporal/activities/submissions.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,SAAS,EAAE,MAAM,kBAAkB,CAAC;AACpD,OAAO,EAAE,IAAI,EAAE,MAAM,WAAW,CAAC;AACjC,OAAO,EAAE,qBAAqB,EAAE,MAAM,kBAAkB,CAAC;AACzD,OAAO,EAAE,eAAe,EAAE,sBAAsB,EAAE,MAAM,cAAc,CAAC;AASvE,sGAAsG;AACtG,MAAM,CAAC,KAAK,UAAU,cAAc,CAClC,eAA2B,EAC3B,YAAwB;IAExB,IAAI,CAAC;QACH,MAAM,UAAU,GAAG,IAAI,CAAC,KAAK,CAAC,MAAM,CAAC,IAAI,CAAC,eAAe,CAAC,CAAC,QAAQ,CAAC,MAAM,CAAC,CAAY,CAAC;QACxF,MAAM,OAAO,GAAG,MAAM,sBAAsB,CAAC,uBAAuB,EAAE,KAAK,EAAE,IAAI,EAAE,EAAE;YACnF,MAAM,OAAO,GAAG,IAAI,CAAC,IAAI,EAAE,oBAAoB,CAAC,CAAC;YACjD,MAAM,QAAQ,GAAG,IAAI,CAAC,IAAI,EAAE,UAAU,CAAC,CAAC;YACxC,MAAM,KAAK,CAAC,QAAQ,CAAC,CAAC;YACtB,MAAM,SAAS,CAAC,OAAO,EAAE,YAAY,CAAC,CAAC;YACvC,MAAM,qBAAqB,CAAC,OAAO,EAAE,QAAQ,CAAC,CAAC;YAC/C,OAAO,MAAM,eAAe,CAAC,QAAQ,CAAC,CAAC;QACzC,CAAC,CAAC,CAAC;QACH,MAAM,MAAM,GAAI,UAA0C,EAAE,MAAM,CAAC;QACnE,OAAO;YACL,MAAM,EAAE,OAAO,MAAM,KAAK,QAAQ,IAAI,MAAM,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,SAAS;YACjE,UAAU;YACV,GAAG,OAAO;SACX,CAAC;IACJ,CAAC;IAAC,MAAM,CAAC;QACP,OAAO,SAAS,CAAC;IACnB,CAAC;AACH,CAAC"}
|
|
@@ -0,0 +1,21 @@
|
|
|
1
|
+
export declare class TaskCompilerInfrastructureError extends Error {
|
|
2
|
+
constructor(message: string, options?: ErrorOptions);
|
|
3
|
+
}
|
|
4
|
+
/**
|
|
5
|
+
* Trusted compiler: unpacks a sandbox submission (definition.json, test.patch, gold.patch),
|
|
6
|
+
* clones the pinned repository on the worker, and renders the native Harbor task bundle.
|
|
7
|
+
* Models never hand-write task.toml, Dockerfiles, or scripts.
|
|
8
|
+
*/
|
|
9
|
+
export interface TaskCompilerInput {
|
|
10
|
+
readonly taskId: string;
|
|
11
|
+
readonly repositoryUrl: string;
|
|
12
|
+
readonly definitionBytes: Uint8Array;
|
|
13
|
+
readonly sourceBundle: Uint8Array;
|
|
14
|
+
readonly token?: string;
|
|
15
|
+
readonly signal?: AbortSignal;
|
|
16
|
+
}
|
|
17
|
+
export interface TaskCompilerServices {
|
|
18
|
+
cloneRepository(repositoryUrl: string, commit: string, destination: string, token?: string, signal?: AbortSignal): Promise<void>;
|
|
19
|
+
}
|
|
20
|
+
export declare function compileSubmittedTask(input: TaskCompilerInput, services?: TaskCompilerServices): Promise<Uint8Array>;
|
|
21
|
+
//# sourceMappingURL=task-compiler.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"task-compiler.d.ts","sourceRoot":"","sources":["../../../src/temporal/activities/task-compiler.ts"],"names":[],"mappings":"AAQA,qBAAa,+BAAgC,SAAQ,KAAK;gBAC5C,OAAO,EAAE,MAAM,EAAE,OAAO,CAAC,EAAE,YAAY;CAIpD;AAED;;;;GAIG;AACH,MAAM,WAAW,iBAAiB;IAChC,QAAQ,CAAC,MAAM,EAAE,MAAM,CAAC;IACxB,QAAQ,CAAC,aAAa,EAAE,MAAM,CAAC;IAC/B,QAAQ,CAAC,eAAe,EAAE,UAAU,CAAC;IACrC,QAAQ,CAAC,YAAY,EAAE,UAAU,CAAC;IAClC,QAAQ,CAAC,KAAK,CAAC,EAAE,MAAM,CAAC;IACxB,QAAQ,CAAC,MAAM,CAAC,EAAE,WAAW,CAAC;CAC/B;AAED,MAAM,WAAW,oBAAoB;IACnC,eAAe,CACb,aAAa,EAAE,MAAM,EACrB,MAAM,EAAE,MAAM,EACd,WAAW,EAAE,MAAM,EACnB,KAAK,CAAC,EAAE,MAAM,EACd,MAAM,CAAC,EAAE,WAAW,GACnB,OAAO,CAAC,IAAI,CAAC,CAAC;CAClB;AAED,wBAAsB,oBAAoB,CACxC,KAAK,EAAE,iBAAiB,EACxB,QAAQ,GAAE,oBAA8C,GACvD,OAAO,CAAC,UAAU,CAAC,CAoCrB"}
|
|
@@ -0,0 +1,63 @@
|
|
|
1
|
+
import { mkdir, readFile, writeFile } from "node:fs/promises";
|
|
2
|
+
import { join } from "node:path";
|
|
3
|
+
import { extractRegularArchive } from "../../archive.js";
|
|
4
|
+
import { taskDefinitionSchema } from "../../contracts.js";
|
|
5
|
+
import { compileHarborTask } from "../../harbor-task.js";
|
|
6
|
+
import { runCommand } from "../../process.js";
|
|
7
|
+
import { withTemporaryDirectory } from "./runtime.js";
|
|
8
|
+
export class TaskCompilerInfrastructureError extends Error {
|
|
9
|
+
constructor(message, options) {
|
|
10
|
+
super(message, options);
|
|
11
|
+
this.name = "TaskCompilerInfrastructureError";
|
|
12
|
+
}
|
|
13
|
+
}
|
|
14
|
+
export async function compileSubmittedTask(input, services = defaultCompilerServices) {
|
|
15
|
+
return await withTemporaryDirectory(`selfbench-compile-${input.taskId}-`, async (root) => {
|
|
16
|
+
const authored = join(root, "authored");
|
|
17
|
+
const repository = join(root, "repository");
|
|
18
|
+
const output = join(root, "harbor-task");
|
|
19
|
+
const archive = join(root, "source-task.tar.gz");
|
|
20
|
+
await mkdir(authored);
|
|
21
|
+
await writeFile(archive, input.sourceBundle);
|
|
22
|
+
await extractRegularArchive(archive, authored, input.signal ? { signal: input.signal } : {});
|
|
23
|
+
await writeFile(join(authored, "definition.json"), input.definitionBytes);
|
|
24
|
+
const definition = taskDefinitionSchema.parse(JSON.parse(Buffer.from(input.definitionBytes).toString("utf8")));
|
|
25
|
+
await services
|
|
26
|
+
.cloneRepository(input.repositoryUrl, definition.baseCommit, repository, input.token, input.signal)
|
|
27
|
+
.catch((error) => {
|
|
28
|
+
throw new TaskCompilerInfrastructureError("failed to materialize source repository", {
|
|
29
|
+
cause: error,
|
|
30
|
+
});
|
|
31
|
+
});
|
|
32
|
+
await compileHarborTask(authored, repository, output);
|
|
33
|
+
const bundle = join(root, "harbor-task.tar.gz");
|
|
34
|
+
await runCommand("tar", ["-czf", bundle, "-C", root, "harbor-task"], input.signal ? { signal: input.signal } : {});
|
|
35
|
+
return await readFile(bundle);
|
|
36
|
+
});
|
|
37
|
+
}
|
|
38
|
+
const defaultCompilerServices = {
|
|
39
|
+
async cloneRepository(repositoryUrl, commit, destination, token, signal) {
|
|
40
|
+
const environment = {
|
|
41
|
+
...process.env,
|
|
42
|
+
...(token ? { GH_TOKEN: token } : {}),
|
|
43
|
+
};
|
|
44
|
+
await withTemporaryDirectory("selfbench-git-auth-", async (root) => {
|
|
45
|
+
if (token) {
|
|
46
|
+
const askpass = join(root, "git-askpass.sh");
|
|
47
|
+
await writeFile(askpass, '#!/bin/sh\ncase "$1" in *Username*) printf x-access-token;; *) printf %s "$GH_TOKEN";; esac\n', { mode: 0o700 });
|
|
48
|
+
environment.GIT_ASKPASS = askpass;
|
|
49
|
+
environment.GIT_TERMINAL_PROMPT = "0";
|
|
50
|
+
}
|
|
51
|
+
await runCommand("git", ["clone", "--no-checkout", "--filter=blob:none", repositoryUrl, destination], { env: environment, ...(signal ? { signal } : {}) });
|
|
52
|
+
await runCommand("git", ["-C", destination, "fetch", "origin", commit], {
|
|
53
|
+
env: environment,
|
|
54
|
+
...(signal ? { signal } : {}),
|
|
55
|
+
});
|
|
56
|
+
await runCommand("git", ["-C", destination, "checkout", "--detach", commit], {
|
|
57
|
+
env: environment,
|
|
58
|
+
...(signal ? { signal } : {}),
|
|
59
|
+
});
|
|
60
|
+
});
|
|
61
|
+
},
|
|
62
|
+
};
|
|
63
|
+
//# sourceMappingURL=task-compiler.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"task-compiler.js","sourceRoot":"","sources":["../../../src/temporal/activities/task-compiler.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,QAAQ,EAAE,SAAS,EAAE,MAAM,kBAAkB,CAAC;AAC9D,OAAO,EAAE,IAAI,EAAE,MAAM,WAAW,CAAC;AACjC,OAAO,EAAE,qBAAqB,EAAE,MAAM,kBAAkB,CAAC;AACzD,OAAO,EAAE,oBAAoB,EAAE,MAAM,oBAAoB,CAAC;AAC1D,OAAO,EAAE,iBAAiB,EAAE,MAAM,sBAAsB,CAAC;AACzD,OAAO,EAAE,UAAU,EAAE,MAAM,kBAAkB,CAAC;AAC9C,OAAO,EAAE,sBAAsB,EAAE,MAAM,cAAc,CAAC;AAEtD,MAAM,OAAO,+BAAgC,SAAQ,KAAK;IACxD,YAAY,OAAe,EAAE,OAAsB;QACjD,KAAK,CAAC,OAAO,EAAE,OAAO,CAAC,CAAC;QACxB,IAAI,CAAC,IAAI,GAAG,iCAAiC,CAAC;IAChD,CAAC;CACF;AA0BD,MAAM,CAAC,KAAK,UAAU,oBAAoB,CACxC,KAAwB,EACxB,WAAiC,uBAAuB;IAExD,OAAO,MAAM,sBAAsB,CAAC,qBAAqB,KAAK,CAAC,MAAM,GAAG,EAAE,KAAK,EAAE,IAAI,EAAE,EAAE;QACvF,MAAM,QAAQ,GAAG,IAAI,CAAC,IAAI,EAAE,UAAU,CAAC,CAAC;QACxC,MAAM,UAAU,GAAG,IAAI,CAAC,IAAI,EAAE,YAAY,CAAC,CAAC;QAC5C,MAAM,MAAM,GAAG,IAAI,CAAC,IAAI,EAAE,aAAa,CAAC,CAAC;QACzC,MAAM,OAAO,GAAG,IAAI,CAAC,IAAI,EAAE,oBAAoB,CAAC,CAAC;QACjD,MAAM,KAAK,CAAC,QAAQ,CAAC,CAAC;QACtB,MAAM,SAAS,CAAC,OAAO,EAAE,KAAK,CAAC,YAAY,CAAC,CAAC;QAC7C,MAAM,qBAAqB,CAAC,OAAO,EAAE,QAAQ,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,MAAM,EAAE,KAAK,CAAC,MAAM,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC;QAC7F,MAAM,SAAS,CAAC,IAAI,CAAC,QAAQ,EAAE,iBAAiB,CAAC,EAAE,KAAK,CAAC,eAAe,CAAC,CAAC;QAE1E,MAAM,UAAU,GAAG,oBAAoB,CAAC,KAAK,CAC3C,IAAI,CAAC,KAAK,CAAC,MAAM,CAAC,IAAI,CAAC,KAAK,CAAC,eAAe,CAAC,CAAC,QAAQ,CAAC,MAAM,CAAC,CAAC,CAChE,CAAC;QACF,MAAM,QAAQ;aACX,eAAe,CACd,KAAK,CAAC,aAAa,EACnB,UAAU,CAAC,UAAU,EACrB,UAAU,EACV,KAAK,CAAC,KAAK,EACX,KAAK,CAAC,MAAM,CACb;aACA,KAAK,CAAC,CAAC,KAAc,EAAE,EAAE;YACxB,MAAM,IAAI,+BAA+B,CAAC,yCAAyC,EAAE;gBACnF,KAAK,EAAE,KAAK;aACb,CAAC,CAAC;QACL,CAAC,CAAC,CAAC;QACL,MAAM,iBAAiB,CAAC,QAAQ,EAAE,UAAU,EAAE,MAAM,CAAC,CAAC;QACtD,MAAM,MAAM,GAAG,IAAI,CAAC,IAAI,EAAE,oBAAoB,CAAC,CAAC;QAChD,MAAM,UAAU,CACd,KAAK,EACL,CAAC,MAAM,EAAE,MAAM,EAAE,IAAI,EAAE,IAAI,EAAE,aAAa,CAAC,EAC3C,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,MAAM,EAAE,KAAK,CAAC,MAAM,EAAE,CAAC,CAAC,CAAC,EAAE,CAC7C,CAAC;QACF,OAAO,MAAM,QAAQ,CAAC,MAAM,CAAC,CAAC;IAChC,CAAC,CAAC,CAAC;AACL,CAAC;AAED,MAAM,uBAAuB,GAAyB;IACpD,KAAK,CAAC,eAAe,CAAC,aAAa,EAAE,MAAM,EAAE,WAAW,EAAE,KAAK,EAAE,MAAM;QACrE,MAAM,WAAW,GAAsB;YACrC,GAAG,OAAO,CAAC,GAAG;YACd,GAAG,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,QAAQ,EAAE,KAAK,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;SACtC,CAAC;QACF,MAAM,sBAAsB,CAAC,qBAAqB,EAAE,KAAK,EAAE,IAAI,EAAE,EAAE;YACjE,IAAI,KAAK,EAAE,CAAC;gBACV,MAAM,OAAO,GAAG,IAAI,CAAC,IAAI,EAAE,gBAAgB,CAAC,CAAC;gBAC7C,MAAM,SAAS,CACb,OAAO,EACP,+FAA+F,EAC/F,EAAE,IAAI,EAAE,KAAK,EAAE,CAChB,CAAC;gBACF,WAAW,CAAC,WAAW,GAAG,OAAO,CAAC;gBAClC,WAAW,CAAC,mBAAmB,GAAG,GAAG,CAAC;YACxC,CAAC;YACD,MAAM,UAAU,CACd,KAAK,EACL,CAAC,OAAO,EAAE,eAAe,EAAE,oBAAoB,EAAE,aAAa,EAAE,WAAW,CAAC,EAC5E,EAAE,GAAG,EAAE,WAAW,EAAE,GAAG,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,MAAM,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,CACpD,CAAC;YACF,MAAM,UAAU,CAAC,KAAK,EAAE,CAAC,IAAI,EAAE,WAAW,EAAE,OAAO,EAAE,QAAQ,EAAE,MAAM,CAAC,EAAE;gBACtE,GAAG,EAAE,WAAW;gBAChB,GAAG,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,MAAM,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;aAC9B,CAAC,CAAC;YACH,MAAM,UAAU,CAAC,KAAK,EAAE,CAAC,IAAI,EAAE,WAAW,EAAE,UAAU,EAAE,UAAU,EAAE,MAAM,CAAC,EAAE;gBAC3E,GAAG,EAAE,WAAW;gBAChB,GAAG,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,MAAM,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;aAC9B,CAAC,CAAC;QACL,CAAC,CAAC,CAAC;IACL,CAAC;CACF,CAAC"}
|
|
@@ -0,0 +1,57 @@
|
|
|
1
|
+
import type { ArtifactRef, AuthoredTask, AuthoredTaskDraft, AuthoringRoundResult, Candidate, Difficulty, DiscoveryResult, ReplayMaterial, ReplayRunRequest, RunRequest, VerifierRoundResult, VerifyOutcome, VerifyStage } from "../../contracts.js";
|
|
2
|
+
export interface DiscoveryShardInput {
|
|
3
|
+
readonly run: RunRequest;
|
|
4
|
+
readonly wave: number;
|
|
5
|
+
readonly shardIndex: number;
|
|
6
|
+
readonly shardCount: number;
|
|
7
|
+
readonly targetCounts: Readonly<Record<Difficulty, number>>;
|
|
8
|
+
readonly excludedSourcePrs: readonly number[];
|
|
9
|
+
}
|
|
10
|
+
/** One authoring agent turn: a fresh session on round 1, a resumed session afterwards. */
|
|
11
|
+
export interface AuthoringRoundInput {
|
|
12
|
+
readonly run: RunRequest;
|
|
13
|
+
readonly candidate: Candidate;
|
|
14
|
+
readonly round: number;
|
|
15
|
+
/** Previous round's pi session file; required for round > 1. */
|
|
16
|
+
readonly session?: ArtifactRef;
|
|
17
|
+
/** Previous round's stored VerifyReport JSON; required for round > 1. */
|
|
18
|
+
readonly report?: ArtifactRef;
|
|
19
|
+
/** In-session verify calls already spent in earlier rounds of this session. */
|
|
20
|
+
readonly verifyCallsUsed?: number;
|
|
21
|
+
/** Read-only verifier suggestions for the next authoring round. */
|
|
22
|
+
readonly feedback?: string;
|
|
23
|
+
}
|
|
24
|
+
/** Trusted compile + audit + Harbor build/smoke/nop/oracle for one submission. */
|
|
25
|
+
export interface CompileAndVerifyInput {
|
|
26
|
+
readonly run: RunRequest;
|
|
27
|
+
readonly candidate: Candidate;
|
|
28
|
+
readonly task: AuthoredTaskDraft;
|
|
29
|
+
readonly stage: VerifyStage;
|
|
30
|
+
readonly round: number;
|
|
31
|
+
}
|
|
32
|
+
/** One verification agent turn over a green task and its latest report. */
|
|
33
|
+
export interface VerifierRoundInput {
|
|
34
|
+
readonly run: RunRequest;
|
|
35
|
+
readonly candidate: Candidate;
|
|
36
|
+
readonly task: AuthoredTask;
|
|
37
|
+
readonly report: ArtifactRef;
|
|
38
|
+
readonly round: number;
|
|
39
|
+
readonly session?: ArtifactRef;
|
|
40
|
+
readonly verifyCallsUsed?: number;
|
|
41
|
+
}
|
|
42
|
+
export interface ExportInput {
|
|
43
|
+
readonly run: RunRequest;
|
|
44
|
+
readonly tasks: readonly AuthoredTask[];
|
|
45
|
+
}
|
|
46
|
+
export interface SelfBenchActivities {
|
|
47
|
+
collectRunProvenance(run: RunRequest): Promise<ArtifactRef>;
|
|
48
|
+
/** Source PRs processed by the listed earlier runs, for cross-run discovery exclusion. */
|
|
49
|
+
collectExcludedSourcePrs(runIds: readonly string[]): Promise<number[]>;
|
|
50
|
+
discoverCandidateShard(input: DiscoveryShardInput): Promise<DiscoveryResult>;
|
|
51
|
+
rebuildReplayCandidates(input: ReplayRunRequest): Promise<ReplayMaterial>;
|
|
52
|
+
runAuthoringRound(input: AuthoringRoundInput): Promise<AuthoringRoundResult>;
|
|
53
|
+
compileAndVerify(input: CompileAndVerifyInput): Promise<VerifyOutcome>;
|
|
54
|
+
runVerifierRound(input: VerifierRoundInput): Promise<VerifierRoundResult>;
|
|
55
|
+
buildExport(input: ExportInput): Promise<ArtifactRef>;
|
|
56
|
+
}
|
|
57
|
+
//# sourceMappingURL=types.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"types.d.ts","sourceRoot":"","sources":["../../../src/temporal/activities/types.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EACV,WAAW,EACX,YAAY,EACZ,iBAAiB,EACjB,oBAAoB,EACpB,SAAS,EACT,UAAU,EACV,eAAe,EACf,cAAc,EACd,gBAAgB,EAChB,UAAU,EACV,mBAAmB,EACnB,aAAa,EACb,WAAW,EACZ,MAAM,oBAAoB,CAAC;AAE5B,MAAM,WAAW,mBAAmB;IAClC,QAAQ,CAAC,GAAG,EAAE,UAAU,CAAC;IACzB,QAAQ,CAAC,IAAI,EAAE,MAAM,CAAC;IACtB,QAAQ,CAAC,UAAU,EAAE,MAAM,CAAC;IAC5B,QAAQ,CAAC,UAAU,EAAE,MAAM,CAAC;IAC5B,QAAQ,CAAC,YAAY,EAAE,QAAQ,CAAC,MAAM,CAAC,UAAU,EAAE,MAAM,CAAC,CAAC,CAAC;IAC5D,QAAQ,CAAC,iBAAiB,EAAE,SAAS,MAAM,EAAE,CAAC;CAC/C;AAED,0FAA0F;AAC1F,MAAM,WAAW,mBAAmB;IAClC,QAAQ,CAAC,GAAG,EAAE,UAAU,CAAC;IACzB,QAAQ,CAAC,SAAS,EAAE,SAAS,CAAC;IAC9B,QAAQ,CAAC,KAAK,EAAE,MAAM,CAAC;IACvB,gEAAgE;IAChE,QAAQ,CAAC,OAAO,CAAC,EAAE,WAAW,CAAC;IAC/B,yEAAyE;IACzE,QAAQ,CAAC,MAAM,CAAC,EAAE,WAAW,CAAC;IAC9B,+EAA+E;IAC/E,QAAQ,CAAC,eAAe,CAAC,EAAE,MAAM,CAAC;IAClC,mEAAmE;IACnE,QAAQ,CAAC,QAAQ,CAAC,EAAE,MAAM,CAAC;CAC5B;AAED,kFAAkF;AAClF,MAAM,WAAW,qBAAqB;IACpC,QAAQ,CAAC,GAAG,EAAE,UAAU,CAAC;IACzB,QAAQ,CAAC,SAAS,EAAE,SAAS,CAAC;IAC9B,QAAQ,CAAC,IAAI,EAAE,iBAAiB,CAAC;IACjC,QAAQ,CAAC,KAAK,EAAE,WAAW,CAAC;IAC5B,QAAQ,CAAC,KAAK,EAAE,MAAM,CAAC;CACxB;AAED,2EAA2E;AAC3E,MAAM,WAAW,kBAAkB;IACjC,QAAQ,CAAC,GAAG,EAAE,UAAU,CAAC;IACzB,QAAQ,CAAC,SAAS,EAAE,SAAS,CAAC;IAC9B,QAAQ,CAAC,IAAI,EAAE,YAAY,CAAC;IAC5B,QAAQ,CAAC,MAAM,EAAE,WAAW,CAAC;IAC7B,QAAQ,CAAC,KAAK,EAAE,MAAM,CAAC;IACvB,QAAQ,CAAC,OAAO,CAAC,EAAE,WAAW,CAAC;IAC/B,QAAQ,CAAC,eAAe,CAAC,EAAE,MAAM,CAAC;CACnC;AAED,MAAM,WAAW,WAAW;IAC1B,QAAQ,CAAC,GAAG,EAAE,UAAU,CAAC;IACzB,QAAQ,CAAC,KAAK,EAAE,SAAS,YAAY,EAAE,CAAC;CACzC;AAED,MAAM,WAAW,mBAAmB;IAClC,oBAAoB,CAAC,GAAG,EAAE,UAAU,GAAG,OAAO,CAAC,WAAW,CAAC,CAAC;IAC5D,0FAA0F;IAC1F,wBAAwB,CAAC,MAAM,EAAE,SAAS,MAAM,EAAE,GAAG,OAAO,CAAC,MAAM,EAAE,CAAC,CAAC;IACvE,sBAAsB,CAAC,KAAK,EAAE,mBAAmB,GAAG,OAAO,CAAC,eAAe,CAAC,CAAC;IAC7E,uBAAuB,CAAC,KAAK,EAAE,gBAAgB,GAAG,OAAO,CAAC,cAAc,CAAC,CAAC;IAC1E,iBAAiB,CAAC,KAAK,EAAE,mBAAmB,GAAG,OAAO,CAAC,oBAAoB,CAAC,CAAC;IAC7E,gBAAgB,CAAC,KAAK,EAAE,qBAAqB,GAAG,OAAO,CAAC,aAAa,CAAC,CAAC;IACvE,gBAAgB,CAAC,KAAK,EAAE,kBAAkB,GAAG,OAAO,CAAC,mBAAmB,CAAC,CAAC;IAC1E,WAAW,CAAC,KAAK,EAAE,WAAW,GAAG,OAAO,CAAC,WAAW,CAAC,CAAC;CACvD"}
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"types.js","sourceRoot":"","sources":["../../../src/temporal/activities/types.ts"],"names":[],"mappings":""}
|
|
@@ -0,0 +1,18 @@
|
|
|
1
|
+
import type { ArtifactStore } from "../../artifacts.js";
|
|
2
|
+
import { type AuthoredTask, type TaskDefinition } from "../../contracts.js";
|
|
3
|
+
import { type CouplingEvidence } from "../../coupling.js";
|
|
4
|
+
export interface VerifierMaterial {
|
|
5
|
+
readonly definition: TaskDefinition;
|
|
6
|
+
readonly instruction: string;
|
|
7
|
+
readonly testPatch: string;
|
|
8
|
+
readonly goldPatch: string;
|
|
9
|
+
readonly couplingEvidence: CouplingEvidence;
|
|
10
|
+
readonly heldOutPaths: readonly string[];
|
|
11
|
+
}
|
|
12
|
+
/**
|
|
13
|
+
* Everything the verification agent judges, gathered from the green bundle on the worker: the
|
|
14
|
+
* instruction, both patches, the environment contract, and the deterministic coupling evidence
|
|
15
|
+
* computed against the exact base snapshot.
|
|
16
|
+
*/
|
|
17
|
+
export declare function buildVerifierMaterial(store: ArtifactStore, task: AuthoredTask): Promise<VerifierMaterial>;
|
|
18
|
+
//# sourceMappingURL=verifier-material.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"verifier-material.d.ts","sourceRoot":"","sources":["../../../src/temporal/activities/verifier-material.ts"],"names":[],"mappings":"AAGA,OAAO,KAAK,EAAE,aAAa,EAAE,MAAM,oBAAoB,CAAC;AACxD,OAAO,EAAE,KAAK,YAAY,EAAE,KAAK,cAAc,EAAwB,MAAM,oBAAoB,CAAC;AAClG,OAAO,EAEL,KAAK,gBAAgB,EAGtB,MAAM,mBAAmB,CAAC;AAK3B,MAAM,WAAW,gBAAgB;IAC/B,QAAQ,CAAC,UAAU,EAAE,cAAc,CAAC;IACpC,QAAQ,CAAC,WAAW,EAAE,MAAM,CAAC;IAC7B,QAAQ,CAAC,SAAS,EAAE,MAAM,CAAC;IAC3B,QAAQ,CAAC,SAAS,EAAE,MAAM,CAAC;IAC3B,QAAQ,CAAC,gBAAgB,EAAE,gBAAgB,CAAC;IAC5C,QAAQ,CAAC,YAAY,EAAE,SAAS,MAAM,EAAE,CAAC;CAC1C;AAED;;;;GAIG;AACH,wBAAsB,qBAAqB,CACzC,KAAK,EAAE,aAAa,EACpB,IAAI,EAAE,YAAY,GACjB,OAAO,CAAC,gBAAgB,CAAC,CAwC3B"}
|
|
@@ -0,0 +1,49 @@
|
|
|
1
|
+
import { mkdir, readFile } from "node:fs/promises";
|
|
2
|
+
import { join } from "node:path";
|
|
3
|
+
import { extractRegularArchive, REPOSITORY_SNAPSHOT_ARCHIVE_OPTIONS } from "../../archive.js";
|
|
4
|
+
import { taskDefinitionSchema } from "../../contracts.js";
|
|
5
|
+
import { buildCouplingEvidence, discoverContractArtifacts, scanBaseContractArtifacts, } from "../../coupling.js";
|
|
6
|
+
import { repositoryRelativePath } from "../../harbor-task/paths.js";
|
|
7
|
+
import { patchPaths } from "../../repair.js";
|
|
8
|
+
import { withTaskBundle } from "./runtime.js";
|
|
9
|
+
/**
|
|
10
|
+
* Everything the verification agent judges, gathered from the green bundle on the worker: the
|
|
11
|
+
* instruction, both patches, the environment contract, and the deterministic coupling evidence
|
|
12
|
+
* computed against the exact base snapshot.
|
|
13
|
+
*/
|
|
14
|
+
export async function buildVerifierMaterial(store, task) {
|
|
15
|
+
return await withTaskBundle(store, task, async (taskDirectory, root) => {
|
|
16
|
+
const [definitionBytes, instruction, testPatch, goldPatch] = await Promise.all([
|
|
17
|
+
store.get(task.definition),
|
|
18
|
+
readFile(join(taskDirectory, "instruction.md"), "utf8"),
|
|
19
|
+
readFile(join(taskDirectory, "tests/test.patch"), "utf8"),
|
|
20
|
+
readFile(join(taskDirectory, "solution/gold.patch"), "utf8"),
|
|
21
|
+
]);
|
|
22
|
+
const definition = taskDefinitionSchema.parse(JSON.parse(Buffer.from(definitionBytes).toString("utf8")));
|
|
23
|
+
const baseDirectory = join(root, "verifier-base");
|
|
24
|
+
await mkdir(baseDirectory);
|
|
25
|
+
await extractRegularArchive(join(taskDirectory, "environment/repo.tar.gz"), baseDirectory, REPOSITORY_SNAPSHOT_ARCHIVE_OPTIONS);
|
|
26
|
+
const candidates = discoverContractArtifacts(testPatch);
|
|
27
|
+
const baseArtifacts = await scanBaseContractArtifacts(baseDirectory, root, candidates);
|
|
28
|
+
const couplingEvidence = buildCouplingEvidence({
|
|
29
|
+
prompt: definition.prompt,
|
|
30
|
+
testPatch,
|
|
31
|
+
goldPatch,
|
|
32
|
+
baseArtifacts,
|
|
33
|
+
});
|
|
34
|
+
return {
|
|
35
|
+
definition,
|
|
36
|
+
instruction,
|
|
37
|
+
testPatch,
|
|
38
|
+
goldPatch,
|
|
39
|
+
couplingEvidence,
|
|
40
|
+
heldOutPaths: [
|
|
41
|
+
...new Set([
|
|
42
|
+
...patchPaths(testPatch),
|
|
43
|
+
...definition.testPaths.map((path) => repositoryRelativePath(definition, path)),
|
|
44
|
+
]),
|
|
45
|
+
],
|
|
46
|
+
};
|
|
47
|
+
});
|
|
48
|
+
}
|
|
49
|
+
//# sourceMappingURL=verifier-material.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"verifier-material.js","sourceRoot":"","sources":["../../../src/temporal/activities/verifier-material.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,KAAK,EAAE,QAAQ,EAAE,MAAM,kBAAkB,CAAC;AACnD,OAAO,EAAE,IAAI,EAAE,MAAM,WAAW,CAAC;AACjC,OAAO,EAAE,qBAAqB,EAAE,mCAAmC,EAAE,MAAM,kBAAkB,CAAC;AAE9F,OAAO,EAA0C,oBAAoB,EAAE,MAAM,oBAAoB,CAAC;AAClG,OAAO,EACL,qBAAqB,EAErB,yBAAyB,EACzB,yBAAyB,GAC1B,MAAM,mBAAmB,CAAC;AAC3B,OAAO,EAAE,sBAAsB,EAAE,MAAM,4BAA4B,CAAC;AACpE,OAAO,EAAE,UAAU,EAAE,MAAM,iBAAiB,CAAC;AAC7C,OAAO,EAAE,cAAc,EAAE,MAAM,cAAc,CAAC;AAW9C;;;;GAIG;AACH,MAAM,CAAC,KAAK,UAAU,qBAAqB,CACzC,KAAoB,EACpB,IAAkB;IAElB,OAAO,MAAM,cAAc,CAAC,KAAK,EAAE,IAAI,EAAE,KAAK,EAAE,aAAa,EAAE,IAAI,EAAE,EAAE;QACrE,MAAM,CAAC,eAAe,EAAE,WAAW,EAAE,SAAS,EAAE,SAAS,CAAC,GAAG,MAAM,OAAO,CAAC,GAAG,CAAC;YAC7E,KAAK,CAAC,GAAG,CAAC,IAAI,CAAC,UAAU,CAAC;YAC1B,QAAQ,CAAC,IAAI,CAAC,aAAa,EAAE,gBAAgB,CAAC,EAAE,MAAM,CAAC;YACvD,QAAQ,CAAC,IAAI,CAAC,aAAa,EAAE,kBAAkB,CAAC,EAAE,MAAM,CAAC;YACzD,QAAQ,CAAC,IAAI,CAAC,aAAa,EAAE,qBAAqB,CAAC,EAAE,MAAM,CAAC;SAC7D,CAAC,CAAC;QACH,MAAM,UAAU,GAAG,oBAAoB,CAAC,KAAK,CAC3C,IAAI,CAAC,KAAK,CAAC,MAAM,CAAC,IAAI,CAAC,eAAe,CAAC,CAAC,QAAQ,CAAC,MAAM,CAAC,CAAC,CAC1D,CAAC;QACF,MAAM,aAAa,GAAG,IAAI,CAAC,IAAI,EAAE,eAAe,CAAC,CAAC;QAClD,MAAM,KAAK,CAAC,aAAa,CAAC,CAAC;QAC3B,MAAM,qBAAqB,CACzB,IAAI,CAAC,aAAa,EAAE,yBAAyB,CAAC,EAC9C,aAAa,EACb,mCAAmC,CACpC,CAAC;QACF,MAAM,UAAU,GAAG,yBAAyB,CAAC,SAAS,CAAC,CAAC;QACxD,MAAM,aAAa,GAAG,MAAM,yBAAyB,CAAC,aAAa,EAAE,IAAI,EAAE,UAAU,CAAC,CAAC;QACvF,MAAM,gBAAgB,GAAG,qBAAqB,CAAC;YAC7C,MAAM,EAAE,UAAU,CAAC,MAAM;YACzB,SAAS;YACT,SAAS;YACT,aAAa;SACd,CAAC,CAAC;QACH,OAAO;YACL,UAAU;YACV,WAAW;YACX,SAAS;YACT,SAAS;YACT,gBAAgB;YAChB,YAAY,EAAE;gBACZ,GAAG,IAAI,GAAG,CAAC;oBACT,GAAG,UAAU,CAAC,SAAS,CAAC;oBACxB,GAAG,UAAU,CAAC,SAAS,CAAC,GAAG,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,sBAAsB,CAAC,UAAU,EAAE,IAAI,CAAC,CAAC;iBAChF,CAAC;aACH;SACF,CAAC;IACJ,CAAC,CAAC,CAAC;AACL,CAAC"}
|
|
@@ -0,0 +1,19 @@
|
|
|
1
|
+
import type { ArtifactStore } from "../../artifacts.js";
|
|
2
|
+
import type { VerifierRoundResult } from "../../contracts.js";
|
|
3
|
+
import { type SandboxRoundResult } from "./round-outcome.js";
|
|
4
|
+
import type { StoredPiSession } from "./runtime.js";
|
|
5
|
+
import type { VerifierRoundInput } from "./types.js";
|
|
6
|
+
export declare const VERDICT_PATH = "/work/verdict/verdict.json";
|
|
7
|
+
export interface VerifierOutcomeInput {
|
|
8
|
+
readonly store: ArtifactStore;
|
|
9
|
+
readonly input: VerifierRoundInput;
|
|
10
|
+
readonly prefix: string;
|
|
11
|
+
/** `<prefix>/attempt-<n>`: where this attempt's pre-decision artifacts go. */
|
|
12
|
+
readonly attemptPrefix: string;
|
|
13
|
+
readonly result: SandboxRoundResult;
|
|
14
|
+
readonly session: StoredPiSession | undefined;
|
|
15
|
+
readonly logUri: string;
|
|
16
|
+
}
|
|
17
|
+
/** Archives a read-only verdict; never imports task modifications from the reviewer. */
|
|
18
|
+
export declare function resolveVerifierOutcome(input: VerifierOutcomeInput): Promise<VerifierRoundResult>;
|
|
19
|
+
//# sourceMappingURL=verifier-outcome.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"verifier-outcome.d.ts","sourceRoot":"","sources":["../../../src/temporal/activities/verifier-outcome.ts"],"names":[],"mappings":"AACA,OAAO,KAAK,EAAE,aAAa,EAAE,MAAM,oBAAoB,CAAC;AACxD,OAAO,KAAK,EAAE,mBAAmB,EAAE,MAAM,oBAAoB,CAAC;AAE9D,OAAO,EAKL,KAAK,kBAAkB,EACxB,MAAM,oBAAoB,CAAC;AAC5B,OAAO,KAAK,EAAE,eAAe,EAAE,MAAM,cAAc,CAAC;AACpD,OAAO,KAAK,EAAE,kBAAkB,EAAE,MAAM,YAAY,CAAC;AAErD,eAAO,MAAM,YAAY,+BAA+B,CAAC;AAezD,MAAM,WAAW,oBAAoB;IACnC,QAAQ,CAAC,KAAK,EAAE,aAAa,CAAC;IAC9B,QAAQ,CAAC,KAAK,EAAE,kBAAkB,CAAC;IACnC,QAAQ,CAAC,MAAM,EAAE,MAAM,CAAC;IACxB,8EAA8E;IAC9E,QAAQ,CAAC,aAAa,EAAE,MAAM,CAAC;IAC/B,QAAQ,CAAC,MAAM,EAAE,kBAAkB,CAAC;IACpC,QAAQ,CAAC,OAAO,EAAE,eAAe,GAAG,SAAS,CAAC;IAC9C,QAAQ,CAAC,MAAM,EAAE,MAAM,CAAC;CACzB;AAED,wFAAwF;AACxF,wBAAsB,sBAAsB,CAC1C,KAAK,EAAE,oBAAoB,GAC1B,OAAO,CAAC,mBAAmB,CAAC,CAsD9B"}
|
|
@@ -0,0 +1,70 @@
|
|
|
1
|
+
import { z } from "zod";
|
|
2
|
+
import { PI_SESSION_OUTPUT_PATH } from "../../pi-session.js";
|
|
3
|
+
import { archiveSandboxResult, classifyRound, piExitCodeFrom, SandboxOutputError, } from "./round-outcome.js";
|
|
4
|
+
export const VERDICT_PATH = "/work/verdict/verdict.json";
|
|
5
|
+
const verdictSchema = z.discriminatedUnion("kind", [
|
|
6
|
+
z.object({ kind: z.literal("accepted"), reason: z.string().min(1) }).passthrough(),
|
|
7
|
+
z
|
|
8
|
+
.object({
|
|
9
|
+
kind: z.literal("suggestions"),
|
|
10
|
+
summary: z.string().min(1),
|
|
11
|
+
suggestions: z.string().min(1),
|
|
12
|
+
})
|
|
13
|
+
.passthrough(),
|
|
14
|
+
z.object({ kind: z.literal("rejected"), reason: z.string().min(1) }),
|
|
15
|
+
z.object({ kind: z.literal("none") }),
|
|
16
|
+
]);
|
|
17
|
+
/** Archives a read-only verdict; never imports task modifications from the reviewer. */
|
|
18
|
+
export async function resolveVerifierOutcome(input) {
|
|
19
|
+
const { store, prefix, attemptPrefix, result, session, logUri } = input;
|
|
20
|
+
const { candidate, round } = input.input;
|
|
21
|
+
const reject = (reason) => ({
|
|
22
|
+
kind: "rejected",
|
|
23
|
+
candidateId: candidate.candidateId,
|
|
24
|
+
reason: `${reason}; log: ${logUri}`,
|
|
25
|
+
});
|
|
26
|
+
const verdictBytes = result.outputs[VERDICT_PATH];
|
|
27
|
+
const missing = await archiveSandboxResult(store, `${attemptPrefix}/sandbox-result.json`, result, [VERDICT_PATH, PI_SESSION_OUTPUT_PATH]);
|
|
28
|
+
const classified = classifyRound({
|
|
29
|
+
round,
|
|
30
|
+
exitCode: result.exitCode,
|
|
31
|
+
piExitCode: piExitCodeFrom(result.stdout),
|
|
32
|
+
missing: missing.filter((path) => path !== PI_SESSION_OUTPUT_PATH),
|
|
33
|
+
sessionCollected: session !== undefined,
|
|
34
|
+
toolCalls: session?.toolCalls ?? [],
|
|
35
|
+
finalMessage: session?.finalMessage,
|
|
36
|
+
providerError: session?.providerError,
|
|
37
|
+
});
|
|
38
|
+
if (classified.kind === "infrastructure") {
|
|
39
|
+
throw new SandboxOutputError(`verifier ${classified.reason}; log: ${logUri}`);
|
|
40
|
+
}
|
|
41
|
+
if (classified.kind === "rejected" || !verdictBytes || !session) {
|
|
42
|
+
return reject(`verifier ${classified.kind === "rejected" ? classified.reason : `round ${round} delivered no verdict`}`);
|
|
43
|
+
}
|
|
44
|
+
const verdict = verdictSchema.safeParse(JSON.parse(Buffer.from(verdictBytes).toString("utf8")));
|
|
45
|
+
if (!verdict.success) {
|
|
46
|
+
return reject(`verifier round ${round} produced an unreadable verdict`);
|
|
47
|
+
}
|
|
48
|
+
await store.put(`${prefix}/verdict.json`, verdictBytes, "application/json");
|
|
49
|
+
if (verdict.data.kind === "none") {
|
|
50
|
+
return reject(`verification agent declined the task${explanation(session)}`);
|
|
51
|
+
}
|
|
52
|
+
if (verdict.data.kind === "rejected")
|
|
53
|
+
return reject(verdict.data.reason);
|
|
54
|
+
if (verdict.data.kind === "accepted") {
|
|
55
|
+
return { kind: "accepted", session: session.ref, reason: verdict.data.reason };
|
|
56
|
+
}
|
|
57
|
+
if (verdict.data.kind === "suggestions") {
|
|
58
|
+
return {
|
|
59
|
+
kind: "suggestions",
|
|
60
|
+
session: session.ref,
|
|
61
|
+
summary: verdict.data.summary,
|
|
62
|
+
suggestions: verdict.data.suggestions,
|
|
63
|
+
};
|
|
64
|
+
}
|
|
65
|
+
return reject("unsupported verifier verdict");
|
|
66
|
+
}
|
|
67
|
+
function explanation(session) {
|
|
68
|
+
return session?.finalMessage ? `: ${session.finalMessage.slice(0, 1_000)}` : "";
|
|
69
|
+
}
|
|
70
|
+
//# sourceMappingURL=verifier-outcome.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"verifier-outcome.js","sourceRoot":"","sources":["../../../src/temporal/activities/verifier-outcome.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,CAAC,EAAE,MAAM,KAAK,CAAC;AAGxB,OAAO,EAAE,sBAAsB,EAAE,MAAM,qBAAqB,CAAC;AAC7D,OAAO,EACL,oBAAoB,EACpB,aAAa,EACb,cAAc,EACd,kBAAkB,GAEnB,MAAM,oBAAoB,CAAC;AAI5B,MAAM,CAAC,MAAM,YAAY,GAAG,4BAA4B,CAAC;AAEzD,MAAM,aAAa,GAAG,CAAC,CAAC,kBAAkB,CAAC,MAAM,EAAE;IACjD,CAAC,CAAC,MAAM,CAAC,EAAE,IAAI,EAAE,CAAC,CAAC,OAAO,CAAC,UAAU,CAAC,EAAE,MAAM,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,WAAW,EAAE;IAClF,CAAC;SACE,MAAM,CAAC;QACN,IAAI,EAAE,CAAC,CAAC,OAAO,CAAC,aAAa,CAAC;QAC9B,OAAO,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC;QAC1B,WAAW,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC;KAC/B,CAAC;SACD,WAAW,EAAE;IAChB,CAAC,CAAC,MAAM,CAAC,EAAE,IAAI,EAAE,CAAC,CAAC,OAAO,CAAC,UAAU,CAAC,EAAE,MAAM,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,CAAC;IACpE,CAAC,CAAC,MAAM,CAAC,EAAE,IAAI,EAAE,CAAC,CAAC,OAAO,CAAC,MAAM,CAAC,EAAE,CAAC;CACtC,CAAC,CAAC;AAaH,wFAAwF;AACxF,MAAM,CAAC,KAAK,UAAU,sBAAsB,CAC1C,KAA2B;IAE3B,MAAM,EAAE,KAAK,EAAE,MAAM,EAAE,aAAa,EAAE,MAAM,EAAE,OAAO,EAAE,MAAM,EAAE,GAAG,KAAK,CAAC;IACxE,MAAM,EAAE,SAAS,EAAE,KAAK,EAAE,GAAG,KAAK,CAAC,KAAK,CAAC;IACzC,MAAM,MAAM,GAAG,CAAC,MAAc,EAAuB,EAAE,CAAC,CAAC;QACvD,IAAI,EAAE,UAAU;QAChB,WAAW,EAAE,SAAS,CAAC,WAAW;QAClC,MAAM,EAAE,GAAG,MAAM,UAAU,MAAM,EAAE;KACpC,CAAC,CAAC;IACH,MAAM,YAAY,GAAG,MAAM,CAAC,OAAO,CAAC,YAAY,CAAC,CAAC;IAClD,MAAM,OAAO,GAAG,MAAM,oBAAoB,CACxC,KAAK,EACL,GAAG,aAAa,sBAAsB,EACtC,MAAM,EACN,CAAC,YAAY,EAAE,sBAAsB,CAAC,CACvC,CAAC;IACF,MAAM,UAAU,GAAG,aAAa,CAAC;QAC/B,KAAK;QACL,QAAQ,EAAE,MAAM,CAAC,QAAQ;QACzB,UAAU,EAAE,cAAc,CAAC,MAAM,CAAC,MAAM,CAAC;QACzC,OAAO,EAAE,OAAO,CAAC,MAAM,CAAC,CAAC,IAAI,EAAE,EAAE,CAAC,IAAI,KAAK,sBAAsB,CAAC;QAClE,gBAAgB,EAAE,OAAO,KAAK,SAAS;QACvC,SAAS,EAAE,OAAO,EAAE,SAAS,IAAI,EAAE;QACnC,YAAY,EAAE,OAAO,EAAE,YAAY;QACnC,aAAa,EAAE,OAAO,EAAE,aAAa;KACtC,CAAC,CAAC;IACH,IAAI,UAAU,CAAC,IAAI,KAAK,gBAAgB,EAAE,CAAC;QACzC,MAAM,IAAI,kBAAkB,CAAC,YAAY,UAAU,CAAC,MAAM,UAAU,MAAM,EAAE,CAAC,CAAC;IAChF,CAAC;IACD,IAAI,UAAU,CAAC,IAAI,KAAK,UAAU,IAAI,CAAC,YAAY,IAAI,CAAC,OAAO,EAAE,CAAC;QAChE,OAAO,MAAM,CACX,YAAY,UAAU,CAAC,IAAI,KAAK,UAAU,CAAC,CAAC,CAAC,UAAU,CAAC,MAAM,CAAC,CAAC,CAAC,SAAS,KAAK,uBAAuB,EAAE,CACzG,CAAC;IACJ,CAAC;IACD,MAAM,OAAO,GAAG,aAAa,CAAC,SAAS,CAAC,IAAI,CAAC,KAAK,CAAC,MAAM,CAAC,IAAI,CAAC,YAAY,CAAC,CAAC,QAAQ,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC;IAChG,IAAI,CAAC,OAAO,CAAC,OAAO,EAAE,CAAC;QACrB,OAAO,MAAM,CAAC,kBAAkB,KAAK,iCAAiC,CAAC,CAAC;IAC1E,CAAC;IACD,MAAM,KAAK,CAAC,GAAG,CAAC,GAAG,MAAM,eAAe,EAAE,YAAY,EAAE,kBAAkB,CAAC,CAAC;IAC5E,IAAI,OAAO,CAAC,IAAI,CAAC,IAAI,KAAK,MAAM,EAAE,CAAC;QACjC,OAAO,MAAM,CAAC,uCAAuC,WAAW,CAAC,OAAO,CAAC,EAAE,CAAC,CAAC;IAC/E,CAAC;IACD,IAAI,OAAO,CAAC,IAAI,CAAC,IAAI,KAAK,UAAU;QAAE,OAAO,MAAM,CAAC,OAAO,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC;IACzE,IAAI,OAAO,CAAC,IAAI,CAAC,IAAI,KAAK,UAAU,EAAE,CAAC;QACrC,OAAO,EAAE,IAAI,EAAE,UAAU,EAAE,OAAO,EAAE,OAAO,CAAC,GAAG,EAAE,MAAM,EAAE,OAAO,CAAC,IAAI,CAAC,MAAM,EAAE,CAAC;IACjF,CAAC;IACD,IAAI,OAAO,CAAC,IAAI,CAAC,IAAI,KAAK,aAAa,EAAE,CAAC;QACxC,OAAO;YACL,IAAI,EAAE,aAAa;YACnB,OAAO,EAAE,OAAO,CAAC,GAAG;YACpB,OAAO,EAAE,OAAO,CAAC,IAAI,CAAC,OAAO;YAC7B,WAAW,EAAE,OAAO,CAAC,IAAI,CAAC,WAAW;SACtC,CAAC;IACJ,CAAC;IACD,OAAO,MAAM,CAAC,8BAA8B,CAAC,CAAC;AAChD,CAAC;AAED,SAAS,WAAW,CAAC,OAAoC;IACvD,OAAO,OAAO,EAAE,YAAY,CAAC,CAAC,CAAC,KAAK,OAAO,CAAC,YAAY,CAAC,KAAK,CAAC,CAAC,EAAE,KAAK,CAAC,EAAE,CAAC,CAAC,CAAC,EAAE,CAAC;AAClF,CAAC"}
|
|
@@ -0,0 +1,8 @@
|
|
|
1
|
+
import type { ArtifactStore } from "../../artifacts.js";
|
|
2
|
+
import type { SelfBenchConfig } from "../../config.js";
|
|
3
|
+
import { type VerifierRoundResult } from "../../contracts.js";
|
|
4
|
+
import type { SandboxExecutor } from "../../sandbox/index.js";
|
|
5
|
+
import type { VerifierRoundInput } from "./types.js";
|
|
6
|
+
/** Fresh read-only review of a mechanically green authoring revision. */
|
|
7
|
+
export declare function runVerifierRound(store: ArtifactStore, sandbox: SandboxExecutor, _harborEnvironment: SelfBenchConfig["harborEnvironment"], input: VerifierRoundInput): Promise<VerifierRoundResult>;
|
|
8
|
+
//# sourceMappingURL=verifier-round.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"verifier-round.d.ts","sourceRoot":"","sources":["../../../src/temporal/activities/verifier-round.ts"],"names":[],"mappings":"AACA,OAAO,KAAK,EAAE,aAAa,EAAE,MAAM,oBAAoB,CAAC;AACxD,OAAO,KAAK,EAAE,eAAe,EAAE,MAAM,iBAAiB,CAAC;AAEvD,OAAO,EACL,KAAK,mBAAmB,EAGzB,MAAM,oBAAoB,CAAC;AAE5B,OAAO,KAAK,EAAE,eAAe,EAAe,MAAM,wBAAwB,CAAC;AAc3E,OAAO,KAAK,EAAE,kBAAkB,EAAE,MAAM,YAAY,CAAC;AAIrD,yEAAyE;AACzE,wBAAsB,gBAAgB,CACpC,KAAK,EAAE,aAAa,EACpB,OAAO,EAAE,eAAe,EACxB,kBAAkB,EAAE,eAAe,CAAC,mBAAmB,CAAC,EACxD,KAAK,EAAE,kBAAkB,GACxB,OAAO,CAAC,mBAAmB,CAAC,CA4G9B"}
|
|
@@ -0,0 +1,110 @@
|
|
|
1
|
+
import { Context } from "@temporalio/activity";
|
|
2
|
+
import { verifierRoundResultSchema, verifyReportSchema, } from "../../contracts.js";
|
|
3
|
+
import { PI_SESSION_OUTPUT_PATH, sessionArtifactKey } from "../../pi-session.js";
|
|
4
|
+
import { loadPiModelAuth } from "../../subscription-auth.js";
|
|
5
|
+
import { renderVerifyReport } from "../../verify-report.js";
|
|
6
|
+
import { withAgentFeed } from "./agent-feed.js";
|
|
7
|
+
import { verifierRoundScript } from "./agent-scripts.js";
|
|
8
|
+
import { AGENT_INACTIVITY_TIMEOUT_MS, VERIFIER_TIMEOUT_MS } from "./constants.js";
|
|
9
|
+
import { verifierPrompt } from "./prompts-verifier.js";
|
|
10
|
+
import { reconcileWrapperStatus, WRAPPER_STATUS_PATH } from "./round-outcome.js";
|
|
11
|
+
import { readAsset, runSandboxWithFailureLog, storePiSession, withActivityHeartbeats, } from "./runtime.js";
|
|
12
|
+
import { buildVerifierMaterial } from "./verifier-material.js";
|
|
13
|
+
import { resolveVerifierOutcome, VERDICT_PATH } from "./verifier-outcome.js";
|
|
14
|
+
/** Fresh read-only review of a mechanically green authoring revision. */
|
|
15
|
+
export async function runVerifierRound(store, sandbox, _harborEnvironment, input) {
|
|
16
|
+
const { run, candidate, task, round } = input;
|
|
17
|
+
const prefix = `runs/${run.runId}/verification/${candidate.candidateId}/round-${round}`;
|
|
18
|
+
// Everything one attempt produces before the round is decided lives under attempt-<n>, so a
|
|
19
|
+
// Temporal retry never collides with the immutable artifacts of the attempt it replaces.
|
|
20
|
+
const attempt = Context.current().info.attempt;
|
|
21
|
+
const attemptPrefix = `${prefix}/attempt-${attempt}`;
|
|
22
|
+
const checkpoint = await store.getByKey(`${prefix}/result.json`);
|
|
23
|
+
if (checkpoint) {
|
|
24
|
+
return verifierRoundResultSchema.parse(JSON.parse(Buffer.from(checkpoint).toString("utf8")));
|
|
25
|
+
}
|
|
26
|
+
Context.current().heartbeat(`verifying ${task.taskId} round ${round}`);
|
|
27
|
+
const [reportBytes, extension, program, piAuth] = await Promise.all([
|
|
28
|
+
store.get(input.report),
|
|
29
|
+
readAsset("dist/extension-verifier.bundle.js"),
|
|
30
|
+
readAsset("dist/sandbox-verifier.bundle.js"),
|
|
31
|
+
loadPiModelAuth(),
|
|
32
|
+
]);
|
|
33
|
+
const report = verifyReportSchema.parse(JSON.parse(Buffer.from(reportBytes).toString("utf8")));
|
|
34
|
+
if (!report.green)
|
|
35
|
+
throw new Error("Read-only verification requires green mechanical checks");
|
|
36
|
+
const material = await buildVerifierMaterial(store, task);
|
|
37
|
+
await store.put(`${attemptPrefix}/coupling-evidence.json`, Buffer.from(JSON.stringify(material.couplingEvidence)), "application/json");
|
|
38
|
+
const prompt = verifierPrompt({
|
|
39
|
+
taskId: task.taskId,
|
|
40
|
+
testSelection: material.definition.testSelection,
|
|
41
|
+
testResults: material.definition.testResults,
|
|
42
|
+
instruction: material.instruction,
|
|
43
|
+
renderedReport: renderVerifyReport(report),
|
|
44
|
+
couplingEvidence: material.couplingEvidence,
|
|
45
|
+
environment: material.definition.environment,
|
|
46
|
+
testPatch: material.testPatch,
|
|
47
|
+
goldPatch: material.goldPatch,
|
|
48
|
+
heldOutPaths: material.heldOutPaths,
|
|
49
|
+
});
|
|
50
|
+
await store.put(`${attemptPrefix}/prompt.md`, Buffer.from(prompt), "text/markdown");
|
|
51
|
+
const logKey = `${attemptPrefix}/sandbox.log`;
|
|
52
|
+
const sandboxResult = await withAgentFeed(store, attemptPrefix, [piAuth.apiKey ?? "", piAuth.authJson ?? ""], (onOutput) => runSandboxWithFailureLog(store, logKey, () => withActivityHeartbeats(`running verifier sandbox for ${task.taskId} round ${round}`, async (options) => sandbox.run({
|
|
53
|
+
runId: run.runId,
|
|
54
|
+
stage: `verify-${candidate.candidateId}-r${round}`,
|
|
55
|
+
timeoutMs: VERIFIER_TIMEOUT_MS,
|
|
56
|
+
inactivityTimeoutMs: AGENT_INACTIVITY_TIMEOUT_MS,
|
|
57
|
+
files: [
|
|
58
|
+
// The bundle (hundreds of MB) is pulled by the sandbox from a signed URL when the
|
|
59
|
+
// store can issue one; otherwise it is loaded inline, never held in a local.
|
|
60
|
+
await bundleFile(store, task.bundle),
|
|
61
|
+
{ path: "/work/sandbox-verifier.js", contents: program },
|
|
62
|
+
{ path: "/work/verifier.js", contents: extension },
|
|
63
|
+
{ path: "/work/prompt.txt", contents: prompt },
|
|
64
|
+
],
|
|
65
|
+
outputPaths: [VERDICT_PATH, PI_SESSION_OUTPUT_PATH, WRAPPER_STATUS_PATH],
|
|
66
|
+
secrets: {
|
|
67
|
+
...(piAuth.apiKey ? { OPENAI_API_KEY: piAuth.apiKey } : {}),
|
|
68
|
+
...(piAuth.authJson ? { SELFBENCH_PI_AUTH_JSON: piAuth.authJson } : {}),
|
|
69
|
+
},
|
|
70
|
+
environment: {
|
|
71
|
+
AUTHOR_MODEL: run.authoring.model,
|
|
72
|
+
AUTHOR_THINKING: run.authoring.reasoningEffort,
|
|
73
|
+
SELFBENCH_TASK_DIRECTORY: "/work/task/harbor-task",
|
|
74
|
+
SELFBENCH_REPO_DIRECTORY: "/work/repo",
|
|
75
|
+
SELFBENCH_VERDICT_OUTPUT: "/work/verdict",
|
|
76
|
+
},
|
|
77
|
+
command: ["bash", "-lc", verifierRoundScript(false)],
|
|
78
|
+
}, { ...options, onOutput }))));
|
|
79
|
+
const result = reconcileWrapperStatus(sandboxResult);
|
|
80
|
+
const log = await store.put(logKey, Buffer.from(`${result.stdout}\n${result.stderr}`), "text/plain");
|
|
81
|
+
const session = await storePiSession(store, sessionArtifactKey(run.runId, "verification", candidate.candidateId, round, attempt), result.outputs[PI_SESSION_OUTPUT_PATH]);
|
|
82
|
+
const outcome = await resolveVerifierOutcome({
|
|
83
|
+
store,
|
|
84
|
+
input,
|
|
85
|
+
prefix,
|
|
86
|
+
attemptPrefix,
|
|
87
|
+
result,
|
|
88
|
+
session,
|
|
89
|
+
logUri: log.uri,
|
|
90
|
+
});
|
|
91
|
+
await store.put(`${prefix}/result.json`, Buffer.from(`${JSON.stringify(outcome, null, 2)}\n`), "application/json");
|
|
92
|
+
return outcome;
|
|
93
|
+
}
|
|
94
|
+
const BUNDLE_URL_TTL_MS = 2 * 60 * 60_000;
|
|
95
|
+
async function bundleFile(store, bundle) {
|
|
96
|
+
try {
|
|
97
|
+
const url = await store.signedReadUrl?.(bundle, BUNDLE_URL_TTL_MS);
|
|
98
|
+
if (url)
|
|
99
|
+
return { path: "/work/task.tar.gz", url, sha256: bundle.sha256 };
|
|
100
|
+
}
|
|
101
|
+
catch (error) {
|
|
102
|
+
// Some ADC credentials can read GCS but cannot sign URLs. Inline fallback keeps verification
|
|
103
|
+
// working; signed URLs remain the fast path for large bundles when a service account is present.
|
|
104
|
+
if (!(error instanceof Error) || !/client_email|sign data|credential/i.test(error.message)) {
|
|
105
|
+
throw error;
|
|
106
|
+
}
|
|
107
|
+
}
|
|
108
|
+
return { path: "/work/task.tar.gz", contents: await store.get(bundle) };
|
|
109
|
+
}
|
|
110
|
+
//# sourceMappingURL=verifier-round.js.map
|