niceeval 0.15.1-canary.76 → 0.16.1
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/INDEX.md +45 -40
- package/dist/adapter-execution-trace.cjs.map +1 -1
- package/dist/adapter-execution-trace.d.cts +26 -0
- package/dist/adapter-execution-trace.d.mts +26 -0
- package/dist/adapter-execution-trace.d.ts +26 -0
- package/dist/adapter-usage.cjs.map +1 -1
- package/dist/adapter-usage.d.cts +9 -0
- package/dist/adapter-usage.d.mts +9 -0
- package/dist/adapter-usage.d.ts +9 -0
- package/dist/adapter.cjs +19 -4
- package/dist/adapter.cjs.map +1 -1
- package/dist/adapter.d.cts +37 -12
- package/dist/adapter.d.mts +37 -12
- package/dist/adapter.d.ts +37 -12
- package/dist/agents/ai-sdk.cjs +3 -0
- package/dist/agents/ai-sdk.cjs.map +1 -1
- package/dist/agents/bub.cjs +3 -0
- package/dist/agents/bub.cjs.map +1 -1
- package/dist/agents/claude-code-native.cjs +28 -2
- package/dist/agents/claude-code-native.cjs.map +1 -1
- package/dist/agents/claude-code.cjs +3 -0
- package/dist/agents/claude-code.cjs.map +1 -1
- package/dist/agents/codex.cjs +3 -0
- package/dist/agents/codex.cjs.map +1 -1
- package/dist/agents/coding-cli-versions.cjs +9 -10
- package/dist/agents/coding-cli-versions.cjs.map +1 -1
- package/dist/agents/coding-cli-versions.d.cts +1 -1
- package/dist/agents/coding-cli-versions.d.mts +1 -1
- package/dist/agents/coding-cli-versions.d.ts +1 -1
- package/dist/agents/deepseek-harness.cjs +3 -0
- package/dist/agents/deepseek-harness.cjs.map +1 -1
- package/dist/agents/hermes.cjs +3 -0
- package/dist/agents/hermes.cjs.map +1 -1
- package/dist/agents/index.cjs.map +1 -1
- package/dist/agents/index.d.cts +3 -2
- package/dist/agents/index.d.mts +3 -2
- package/dist/agents/index.d.ts +3 -2
- package/dist/agents/omp.cjs +3 -0
- package/dist/agents/omp.cjs.map +1 -1
- package/dist/agents/openai-compat.cjs +6 -0
- package/dist/agents/openai-compat.cjs.map +1 -1
- package/dist/agents/openclaw.cjs +3 -0
- package/dist/agents/openclaw.cjs.map +1 -1
- package/dist/agents/opencode.cjs +3 -0
- package/dist/agents/opencode.cjs.map +1 -1
- package/dist/agents/sdk-streams.cjs +9 -0
- package/dist/agents/sdk-streams.cjs.map +1 -1
- package/dist/agents/ui-message-stream.cjs +3 -0
- package/dist/agents/ui-message-stream.cjs.map +1 -1
- package/dist/assertions/api.cjs.map +1 -1
- package/dist/assertions/api.d.cts +27 -5
- package/dist/assertions/api.d.mts +27 -5
- package/dist/assertions/api.d.ts +27 -5
- package/dist/assertions/collection.cjs +6 -0
- package/dist/assertions/collection.cjs.map +1 -1
- package/dist/assertions/collection.d.cts +4 -2
- package/dist/assertions/collection.d.mts +4 -2
- package/dist/assertions/collection.d.ts +4 -2
- package/dist/assertions/collection.js +12 -8
- package/dist/assertions/context-match.cjs +70 -0
- package/dist/assertions/context-match.cjs.map +1 -0
- package/dist/assertions/context-match.d.cts +107 -0
- package/dist/assertions/context-match.d.mts +107 -0
- package/dist/assertions/context-match.d.ts +107 -0
- package/dist/assertions/context-match.js +12 -0
- package/dist/assertions/context-match.js.map +1 -0
- package/dist/assertions/full-content.cjs +109 -0
- package/dist/assertions/full-content.cjs.map +1 -0
- package/dist/assertions/full-content.d.cts +6 -0
- package/dist/assertions/full-content.d.mts +6 -0
- package/dist/assertions/full-content.d.ts +6 -0
- package/dist/assertions/full-content.js +12 -0
- package/dist/assertions/full-content.js.map +1 -0
- package/dist/assertions/judge-presets.cjs +2 -34
- package/dist/assertions/judge-presets.cjs.map +1 -1
- package/dist/assertions/judge-presets.d.cts +6 -9
- package/dist/assertions/judge-presets.d.mts +6 -9
- package/dist/assertions/judge-presets.d.ts +6 -9
- package/dist/assertions/judge-presets.js +8 -8
- package/dist/assertions/judge.cjs +107 -29
- package/dist/assertions/judge.cjs.map +1 -1
- package/dist/assertions/judge.d.cts +2 -0
- package/dist/assertions/judge.d.mts +2 -0
- package/dist/assertions/judge.d.ts +2 -0
- package/dist/assertions/match.cjs +244 -5
- package/dist/assertions/match.cjs.map +1 -1
- package/dist/assertions/match.d.cts +43 -3
- package/dist/assertions/match.d.mts +43 -3
- package/dist/assertions/match.d.ts +43 -3
- package/dist/assertions/match.js +88 -78
- package/dist/assertions/material.cjs +520 -0
- package/dist/assertions/material.cjs.map +1 -0
- package/dist/assertions/material.d.cts +56 -0
- package/dist/assertions/material.d.mts +56 -0
- package/dist/assertions/material.d.ts +56 -0
- package/dist/assertions/material.js +20 -0
- package/dist/assertions/material.js.map +1 -0
- package/dist/assertions/record/attachment.cjs +5 -2
- package/dist/assertions/record/attachment.cjs.map +1 -1
- package/dist/assertions/record/codec.cjs +2 -1
- package/dist/assertions/record/codec.cjs.map +1 -1
- package/dist/assertions/record/model.cjs.map +1 -1
- package/dist/assertions/record/model.d.cts +6 -1
- package/dist/assertions/record/model.d.mts +6 -1
- package/dist/assertions/record/model.d.ts +6 -1
- package/dist/assertions/runtime.cjs +207 -14
- package/dist/assertions/runtime.cjs.map +1 -1
- package/dist/assertions/runtime.d.cts +9 -0
- package/dist/assertions/runtime.d.mts +9 -0
- package/dist/assertions/runtime.d.ts +9 -0
- package/dist/assertions/runtime.js +14 -8
- package/dist/assertions/score-match-audit.cjs +18 -8
- package/dist/assertions/score-match-audit.cjs.map +1 -1
- package/dist/assertions/score-match-gateway.cjs +83 -25
- package/dist/assertions/score-match-gateway.cjs.map +1 -1
- package/dist/assertions/score-match-gateway.d.cts +2 -0
- package/dist/assertions/score-match-gateway.d.mts +2 -0
- package/dist/assertions/score-match-gateway.d.ts +2 -0
- package/dist/assertions/value-composition.test-d.cjs +77 -0
- package/dist/assertions/value-composition.test-d.cjs.map +1 -0
- package/dist/assertions/value-composition.test-d.d.cts +1 -0
- package/dist/assertions/value-composition.test-d.d.mts +1 -0
- package/dist/assertions/value-composition.test-d.d.ts +1 -0
- package/dist/assertions/value-composition.test-d.js +4 -0
- package/dist/assertions/value-composition.test-d.js.map +1 -0
- package/dist/context/assert-first.cjs +207 -213
- package/dist/context/assert-first.cjs.map +1 -1
- package/dist/context/assert-first.d.cts +49 -23
- package/dist/context/assert-first.d.mts +49 -23
- package/dist/context/assert-first.d.ts +49 -23
- package/dist/context/session.cjs +111 -9
- package/dist/context/session.cjs.map +1 -1
- package/dist/context/session.d.cts +18 -0
- package/dist/context/session.d.mts +18 -0
- package/dist/context/session.d.ts +18 -0
- package/dist/define.cjs +17 -1
- package/dist/define.cjs.map +1 -1
- package/dist/eval/cli/contribution.cjs +2 -0
- package/dist/eval/cli/contribution.cjs.map +1 -1
- package/dist/eval/host/index.cjs +2 -0
- package/dist/eval/host/index.cjs.map +1 -1
- package/dist/eval/record/evaluation.cjs +3 -0
- package/dist/eval/record/evaluation.cjs.map +1 -1
- package/dist/eval/record/membership-provenance.d.cts +15 -15
- package/dist/eval/record/membership-provenance.d.mts +15 -15
- package/dist/eval/record/membership-provenance.d.ts +15 -15
- package/dist/eval/record/score.cjs +7 -0
- package/dist/eval/record/score.cjs.map +1 -1
- package/dist/eval/record/sealed-assertion.cjs +3 -0
- package/dist/eval/record/sealed-assertion.cjs.map +1 -1
- package/dist/eval/record/verdict.cjs +7 -0
- package/dist/eval/record/verdict.cjs.map +1 -1
- package/dist/expect/index.cjs +8 -2
- package/dist/expect/index.cjs.map +1 -1
- package/dist/expect/index.d.cts +9 -4
- package/dist/expect/index.d.mts +9 -4
- package/dist/expect/index.d.ts +9 -4
- package/dist/expect/index.js +72 -62
- package/dist/expect/index.mjs +72 -62
- package/dist/experiment/host/cli/contribution.cjs +37 -23
- package/dist/experiment/host/cli/contribution.cjs.map +1 -1
- package/dist/experiment/host/cli/contribution.d.cts +9 -1
- package/dist/experiment/host/cli/contribution.d.mts +9 -1
- package/dist/experiment/host/cli/contribution.d.ts +9 -1
- package/dist/experiment/host/current.cjs +74 -0
- package/dist/experiment/host/current.cjs.map +1 -0
- package/dist/experiment/host/current.d.cts +33 -0
- package/dist/experiment/host/current.d.mts +33 -0
- package/dist/experiment/host/current.d.ts +33 -0
- package/dist/experiment/host/current.js +10 -0
- package/dist/experiment/host/current.js.map +1 -0
- package/dist/experiment/host/index.cjs +17 -174
- package/dist/experiment/host/index.cjs.map +1 -1
- package/dist/experiment/host/index.d.cts +8 -54
- package/dist/experiment/host/index.d.mts +8 -54
- package/dist/experiment/host/index.d.ts +8 -54
- package/dist/experiment/host/operations.cjs +33 -10
- package/dist/experiment/host/operations.cjs.map +1 -1
- package/dist/experiment/host/operations.d.cts +29 -1
- package/dist/experiment/host/operations.d.mts +29 -1
- package/dist/experiment/host/operations.d.ts +29 -1
- package/dist/experiment/host/operations.js +6 -4
- package/dist/experiment/host/requirements.cjs +3 -0
- package/dist/experiment/host/requirements.cjs.map +1 -0
- package/dist/experiment/host/requirements.d.cts +3 -0
- package/dist/experiment/host/requirements.d.mts +3 -0
- package/dist/experiment/host/requirements.d.ts +3 -0
- package/dist/experiment/host/requirements.js +4 -0
- package/dist/experiment/host/requirements.js.map +1 -0
- package/dist/experiment/host/runtime.cjs +179 -0
- package/dist/experiment/host/runtime.cjs.map +1 -0
- package/dist/experiment/host/runtime.d.cts +79 -0
- package/dist/experiment/host/runtime.d.mts +79 -0
- package/dist/experiment/host/runtime.d.ts +79 -0
- package/dist/experiment/host/runtime.js +6 -0
- package/dist/experiment/host/runtime.js.map +1 -0
- package/dist/experiment/host/types.cjs.map +1 -1
- package/dist/experiment/host/types.d.cts +17 -16
- package/dist/experiment/host/types.d.mts +17 -16
- package/dist/experiment/host/types.d.ts +17 -16
- package/dist/fresh-import.cjs +2 -1
- package/dist/fresh-import.cjs.map +1 -1
- package/dist/index.cjs +4 -2
- package/dist/index.cjs.map +1 -1
- package/dist/index.d.cts +12 -6
- package/dist/index.d.mts +12 -6
- package/dist/index.d.ts +12 -6
- package/dist/index.js +32 -30
- package/dist/index.mjs +32 -30
- package/dist/inspection/artifacts.cjs +83 -54
- package/dist/inspection/artifacts.cjs.map +1 -1
- package/dist/inspection/artifacts.d.cts +8 -0
- package/dist/inspection/artifacts.d.mts +8 -0
- package/dist/inspection/artifacts.d.ts +8 -0
- package/dist/inspection/artifacts.js +2 -0
- package/dist/inspection/assertions.cjs +4 -1
- package/dist/inspection/assertions.cjs.map +1 -1
- package/dist/inspection/cli/contribution.cjs +27 -4
- package/dist/inspection/cli/contribution.cjs.map +1 -1
- package/dist/inspection/cli/contribution.d.cts +2 -1
- package/dist/inspection/cli/contribution.d.mts +2 -1
- package/dist/inspection/cli/contribution.d.ts +2 -1
- package/dist/inspection/execution-traces.cjs +114 -8
- package/dist/inspection/execution-traces.cjs.map +1 -1
- package/dist/inspection/execution-traces.d.cts +38 -0
- package/dist/inspection/execution-traces.d.mts +38 -0
- package/dist/inspection/execution-traces.d.ts +38 -0
- package/dist/inspection/facts.cjs +12 -0
- package/dist/inspection/facts.cjs.map +1 -1
- package/dist/inspection/limits.cjs +5 -1
- package/dist/inspection/limits.cjs.map +1 -1
- package/dist/inspection/limits.d.cts +4 -0
- package/dist/inspection/limits.d.mts +4 -0
- package/dist/inspection/limits.d.ts +4 -0
- package/dist/inspection/limits.js +6 -2
- package/dist/inspection/overview.cjs +74 -33
- package/dist/inspection/overview.cjs.map +1 -1
- package/dist/inspection/overview.d.cts +19 -1
- package/dist/inspection/overview.d.mts +19 -1
- package/dist/inspection/overview.d.ts +19 -1
- package/dist/inspection/overview.js +4 -2
- package/dist/inspection/project-input.cjs +9 -0
- package/dist/inspection/project-input.cjs.map +1 -0
- package/dist/inspection/project-input.d.cts +20 -0
- package/dist/inspection/project-input.d.mts +20 -0
- package/dist/inspection/project-input.d.ts +20 -0
- package/dist/inspection/project-input.js +8 -0
- package/dist/inspection/project-input.js.map +1 -0
- package/dist/inspection/project-result.cjs +60 -0
- package/dist/inspection/project-result.cjs.map +1 -0
- package/dist/inspection/project-result.d.cts +1264 -0
- package/dist/inspection/project-result.d.mts +1264 -0
- package/dist/inspection/project-result.d.ts +1264 -0
- package/dist/inspection/project-result.js +10 -0
- package/dist/inspection/project-result.js.map +1 -0
- package/dist/inspection/project-source.cjs +109 -0
- package/dist/inspection/project-source.cjs.map +1 -0
- package/dist/inspection/project-source.d.cts +10 -0
- package/dist/inspection/project-source.d.mts +10 -0
- package/dist/inspection/project-source.d.ts +10 -0
- package/dist/inspection/project-source.js +6 -0
- package/dist/inspection/project-source.js.map +1 -0
- package/dist/inspection/project.cjs +100 -0
- package/dist/inspection/project.cjs.map +1 -0
- package/dist/inspection/project.d.cts +3 -0
- package/dist/inspection/project.d.mts +3 -0
- package/dist/inspection/project.d.ts +3 -0
- package/dist/inspection/project.js +6 -0
- package/dist/inspection/project.js.map +1 -0
- package/dist/inspection/protocol.cjs +19 -9
- package/dist/inspection/protocol.cjs.map +1 -1
- package/dist/inspection/protocol.d.cts +5506 -38189
- package/dist/inspection/protocol.d.mts +5506 -38189
- package/dist/inspection/protocol.d.ts +5506 -38189
- package/dist/inspection/public.mjs +896 -84
- package/dist/inspection/query.cjs +2 -0
- package/dist/inspection/query.cjs.map +1 -1
- package/dist/inspection/results.cjs +234 -29
- package/dist/inspection/results.cjs.map +1 -1
- package/dist/inspection/results.d.cts +870 -13
- package/dist/inspection/results.d.mts +870 -13
- package/dist/inspection/results.d.ts +870 -13
- package/dist/inspection/results.js +36 -34
- package/dist/inspection/select.cjs +126 -14
- package/dist/inspection/select.cjs.map +1 -1
- package/dist/inspection/select.d.cts +1 -1
- package/dist/inspection/select.d.mts +1 -1
- package/dist/inspection/select.d.ts +1 -1
- package/dist/inspection/source.cjs +4 -0
- package/dist/inspection/source.cjs.map +1 -1
- package/dist/inspection/sources.cjs +5 -0
- package/dist/inspection/sources.cjs.map +1 -1
- package/dist/inspection/trace.cjs +82 -84
- package/dist/inspection/trace.cjs.map +1 -1
- package/dist/inspection/trace.d.cts +7 -2
- package/dist/inspection/trace.d.mts +7 -2
- package/dist/inspection/trace.d.ts +7 -2
- package/dist/model-slots.cjs +35 -0
- package/dist/model-slots.cjs.map +1 -0
- package/dist/model-slots.d.cts +19 -0
- package/dist/model-slots.d.mts +19 -0
- package/dist/model-slots.d.ts +19 -0
- package/dist/model-slots.js +8 -0
- package/dist/model-slots.js.map +1 -0
- package/dist/o11y/adapter-call-price.cjs +4 -0
- package/dist/o11y/adapter-call-price.cjs.map +1 -1
- package/dist/o11y/adapter-call-price.d.cts +24 -0
- package/dist/o11y/adapter-call-price.d.mts +24 -0
- package/dist/o11y/adapter-call-price.d.ts +24 -0
- package/dist/o11y/adapter-call-price.js +6 -2
- package/dist/o11y/adapter-usage-projection.cjs +158 -0
- package/dist/o11y/adapter-usage-projection.cjs.map +1 -0
- package/dist/o11y/adapter-usage-projection.d.cts +141 -0
- package/dist/o11y/adapter-usage-projection.d.mts +141 -0
- package/dist/o11y/adapter-usage-projection.d.ts +141 -0
- package/dist/o11y/adapter-usage-projection.js +20 -0
- package/dist/o11y/adapter-usage-projection.js.map +1 -0
- package/dist/o11y/eval-usage.cjs +149 -0
- package/dist/o11y/eval-usage.cjs.map +1 -0
- package/dist/o11y/eval-usage.d.cts +46 -0
- package/dist/o11y/eval-usage.d.mts +46 -0
- package/dist/o11y/eval-usage.d.ts +46 -0
- package/dist/o11y/eval-usage.js +8 -0
- package/dist/o11y/eval-usage.js.map +1 -0
- package/dist/o11y/judge-call-price.cjs +62 -0
- package/dist/o11y/judge-call-price.cjs.map +1 -0
- package/dist/o11y/judge-call-price.d.cts +3 -0
- package/dist/o11y/judge-call-price.d.mts +3 -0
- package/dist/o11y/judge-call-price.d.ts +3 -0
- package/dist/o11y/judge-call-price.js +6 -0
- package/dist/o11y/judge-call-price.js.map +1 -0
- package/dist/o11y/judge-usage-projection.cjs +134 -0
- package/dist/o11y/judge-usage-projection.cjs.map +1 -0
- package/dist/o11y/judge-usage-projection.d.cts +18 -0
- package/dist/o11y/judge-usage-projection.d.mts +18 -0
- package/dist/o11y/judge-usage-projection.d.ts +18 -0
- package/dist/o11y/judge-usage-projection.js +20 -0
- package/dist/o11y/judge-usage-projection.js.map +1 -0
- package/dist/o11y/judge-usage-receipt.cjs +135 -0
- package/dist/o11y/judge-usage-receipt.cjs.map +1 -0
- package/dist/o11y/judge-usage-receipt.d.cts +5 -0
- package/dist/o11y/judge-usage-receipt.d.mts +5 -0
- package/dist/o11y/judge-usage-receipt.d.ts +5 -0
- package/dist/o11y/judge-usage-receipt.js +6 -0
- package/dist/o11y/judge-usage-receipt.js.map +1 -0
- package/dist/o11y/judge-usage.cjs +171 -0
- package/dist/o11y/judge-usage.cjs.map +1 -0
- package/dist/o11y/judge-usage.d.cts +35 -0
- package/dist/o11y/judge-usage.d.mts +35 -0
- package/dist/o11y/judge-usage.d.ts +35 -0
- package/dist/o11y/judge-usage.js +6 -0
- package/dist/o11y/judge-usage.js.map +1 -0
- package/dist/plugin/contracts.cjs +2 -0
- package/dist/plugin/contracts.cjs.map +1 -1
- package/dist/plugin/link.cjs +7 -0
- package/dist/plugin/link.cjs.map +1 -1
- package/dist/record/attachment/protocol.cjs +8 -6
- package/dist/record/attachment/protocol.cjs.map +1 -1
- package/dist/record/attachment/protocol.d.cts +10 -0
- package/dist/record/attachment/protocol.d.mts +10 -0
- package/dist/record/attachment/protocol.d.ts +10 -0
- package/dist/record/attachment/protocol.js +38 -36
- package/dist/record/codec/seal-manifest.d.cts +3 -3
- package/dist/record/codec/seal-manifest.d.mts +3 -3
- package/dist/record/codec/seal-manifest.d.ts +3 -3
- package/dist/record/family/adapter-usage/definition.d.cts +1 -0
- package/dist/record/family/adapter-usage/definition.d.mts +1 -0
- package/dist/record/family/adapter-usage/definition.d.ts +1 -0
- package/dist/record/family/adapter-usage/persistence.cjs +1 -1
- package/dist/record/family/adapter-usage/persistence.cjs.map +1 -1
- package/dist/record/family/adapter-usage/persistence.d.cts +2 -1
- package/dist/record/family/adapter-usage/persistence.d.mts +2 -1
- package/dist/record/family/adapter-usage/persistence.d.ts +2 -1
- package/dist/record/family/adapter-usage/schema.cjs +21 -2
- package/dist/record/family/adapter-usage/schema.cjs.map +1 -1
- package/dist/record/family/adapter-usage/schema.d.cts +85 -0
- package/dist/record/family/adapter-usage/schema.d.mts +85 -0
- package/dist/record/family/adapter-usage/schema.d.ts +85 -0
- package/dist/record/family/adapter-usage/schema.js +18 -12
- package/dist/record/family/catalog.cjs +31 -27
- package/dist/record/family/catalog.cjs.map +1 -1
- package/dist/record/family/catalog.d.cts +94 -3
- package/dist/record/family/catalog.d.mts +94 -3
- package/dist/record/family/catalog.d.ts +94 -3
- package/dist/record/family/current.cjs +45 -42
- package/dist/record/family/current.cjs.map +1 -1
- package/dist/record/family/current.d.cts +240 -1
- package/dist/record/family/current.d.mts +240 -1
- package/dist/record/family/current.d.ts +240 -1
- package/dist/record/family/execution-traces/definition.d.cts +26 -0
- package/dist/record/family/execution-traces/definition.d.mts +26 -0
- package/dist/record/family/execution-traces/definition.d.ts +26 -0
- package/dist/record/family/execution-traces/definition.js +20 -14
- package/dist/record/family/execution-traces/schema.cjs +57 -1
- package/dist/record/family/execution-traces/schema.cjs.map +1 -1
- package/dist/record/family/execution-traces/schema.d.cts +114 -0
- package/dist/record/family/execution-traces/schema.d.mts +114 -0
- package/dist/record/family/execution-traces/schema.d.ts +114 -0
- package/dist/record/family/execution-traces/schema.js +18 -12
- package/dist/record/family/judge-usage/definition.cjs +12 -0
- package/dist/record/family/judge-usage/definition.cjs.map +1 -0
- package/dist/record/family/judge-usage/definition.d.cts +91 -0
- package/dist/record/family/judge-usage/definition.d.mts +91 -0
- package/dist/record/family/judge-usage/definition.d.ts +91 -0
- package/dist/record/family/judge-usage/definition.js +6 -0
- package/dist/record/family/judge-usage/definition.js.map +1 -0
- package/dist/record/family/judge-usage/persistence.cjs +10 -0
- package/dist/record/family/judge-usage/persistence.cjs.map +1 -0
- package/dist/record/family/judge-usage/persistence.d.cts +91 -0
- package/dist/record/family/judge-usage/persistence.d.mts +91 -0
- package/dist/record/family/judge-usage/persistence.d.ts +91 -0
- package/dist/record/family/judge-usage/persistence.js +6 -0
- package/dist/record/family/judge-usage/persistence.js.map +1 -0
- package/dist/record/family/judge-usage/schema.cjs +307 -0
- package/dist/record/family/judge-usage/schema.cjs.map +1 -0
- package/dist/record/family/judge-usage/schema.d.cts +351 -0
- package/dist/record/family/judge-usage/schema.d.mts +351 -0
- package/dist/record/family/judge-usage/schema.d.ts +351 -0
- package/dist/record/family/judge-usage/schema.js +18 -0
- package/dist/record/family/judge-usage/schema.js.map +1 -0
- package/dist/record/host/sqlite-host.cjs +55 -9
- package/dist/record/host/sqlite-host.cjs.map +1 -1
- package/dist/record/host/types.cjs.map +1 -1
- package/dist/record/host/types.d.cts +7 -0
- package/dist/record/host/types.d.mts +7 -0
- package/dist/record/host/types.d.ts +7 -0
- package/dist/record/model/run-context.cjs +7 -1
- package/dist/record/model/run-context.cjs.map +1 -1
- package/dist/record/model/run-context.d.cts +4 -0
- package/dist/record/model/run-context.d.mts +4 -0
- package/dist/record/model/run-context.d.ts +4 -0
- package/dist/record/model/seal-manifest.d.cts +1 -1
- package/dist/record/model/seal-manifest.d.mts +1 -1
- package/dist/record/model/seal-manifest.d.ts +1 -1
- package/dist/record/sqlite/client.cjs +18 -5
- package/dist/record/sqlite/client.cjs.map +1 -1
- package/dist/record/sqlite/client.d.cts +1 -1
- package/dist/record/sqlite/client.d.mts +1 -1
- package/dist/record/sqlite/client.d.ts +1 -1
- package/dist/record/sqlite/project-state-database.cjs +29 -8
- package/dist/record/sqlite/project-state-database.cjs.map +1 -1
- package/dist/run/host/runtime.cjs +4 -0
- package/dist/run/host/runtime.cjs.map +1 -1
- package/dist/run/storage/sqlite.cjs +20 -0
- package/dist/run/storage/sqlite.cjs.map +1 -1
- package/dist/runner/adapter-execution-trace.cjs +125 -2
- package/dist/runner/adapter-execution-trace.cjs.map +1 -1
- package/dist/runner/adapter-execution-trace.d.cts +1 -1
- package/dist/runner/adapter-execution-trace.d.mts +1 -1
- package/dist/runner/adapter-execution-trace.d.ts +1 -1
- package/dist/runner/adapter-usage.cjs +55 -7
- package/dist/runner/adapter-usage.cjs.map +1 -1
- package/dist/runner/adapter-usage.d.cts +9 -2
- package/dist/runner/adapter-usage.d.mts +9 -2
- package/dist/runner/adapter-usage.d.ts +9 -2
- package/dist/runner/adoption.cjs +9 -0
- package/dist/runner/adoption.cjs.map +1 -1
- package/dist/runner/attempt-cancellation.cjs +12 -0
- package/dist/runner/attempt-cancellation.cjs.map +1 -0
- package/dist/runner/attempt-cancellation.d.cts +6 -0
- package/dist/runner/attempt-cancellation.d.mts +6 -0
- package/dist/runner/attempt-cancellation.d.ts +6 -0
- package/dist/runner/attempt-cancellation.js +6 -0
- package/dist/runner/attempt-cancellation.js.map +1 -0
- package/dist/runner/attempt-resources.cjs +10 -4
- package/dist/runner/attempt-resources.cjs.map +1 -1
- package/dist/runner/attempt-resources.d.cts +2 -1
- package/dist/runner/attempt-resources.d.mts +2 -1
- package/dist/runner/attempt-resources.d.ts +2 -1
- package/dist/runner/attempt.cjs +1177 -1139
- package/dist/runner/attempt.cjs.map +1 -1
- package/dist/runner/config-identity.cjs +20 -1
- package/dist/runner/config-identity.cjs.map +1 -1
- package/dist/runner/config-identity.d.cts +5 -0
- package/dist/runner/config-identity.d.mts +5 -0
- package/dist/runner/config-identity.d.ts +5 -0
- package/dist/runner/discover.cjs +14 -0
- package/dist/runner/discover.cjs.map +1 -1
- package/dist/runner/discover.d.cts +3 -0
- package/dist/runner/discover.d.mts +3 -0
- package/dist/runner/discover.d.ts +3 -0
- package/dist/runner/feedback/human.cjs +9 -4
- package/dist/runner/feedback/human.cjs.map +1 -1
- package/dist/runner/fingerprint.cjs +1 -0
- package/dist/runner/fingerprint.cjs.map +1 -1
- package/dist/runner/judge-usage.cjs +18 -0
- package/dist/runner/judge-usage.cjs.map +1 -0
- package/dist/runner/judge-usage.d.cts +5 -0
- package/dist/runner/judge-usage.d.mts +5 -0
- package/dist/runner/judge-usage.d.ts +5 -0
- package/dist/runner/judge-usage.js +8 -0
- package/dist/runner/judge-usage.js.map +1 -0
- package/dist/runner/project-target.cjs.map +1 -1
- package/dist/runner/project-target.d.cts +2 -0
- package/dist/runner/project-target.d.mts +2 -0
- package/dist/runner/project-target.d.ts +2 -0
- package/dist/runner/record/planning.cjs +1 -0
- package/dist/runner/record/planning.cjs.map +1 -1
- package/dist/runner/record.cjs +4 -0
- package/dist/runner/record.cjs.map +1 -1
- package/dist/runner/reporters/braintrust.cjs +2 -0
- package/dist/runner/reporters/braintrust.cjs.map +1 -1
- package/dist/runner/reuse-diagnostics.cjs +9 -10
- package/dist/runner/reuse-diagnostics.cjs.map +1 -1
- package/dist/runner/reuse-diagnostics.d.cts +5 -6
- package/dist/runner/reuse-diagnostics.d.mts +5 -6
- package/dist/runner/reuse-diagnostics.d.ts +5 -6
- package/dist/runner/reuse-diagnostics.js +4 -4
- package/dist/runner/reuse-plan.cjs +100 -48
- package/dist/runner/reuse-plan.cjs.map +1 -1
- package/dist/runner/reuse-plan.d.cts +60 -3
- package/dist/runner/reuse-plan.d.mts +60 -3
- package/dist/runner/reuse-plan.d.ts +60 -3
- package/dist/runner/reuse-plan.js +14 -10
- package/dist/runner/run.cjs +8 -6
- package/dist/runner/run.cjs.map +1 -1
- package/dist/runner/session.cjs +1 -1
- package/dist/runner/session.cjs.map +1 -1
- package/dist/runner/session.d.cts +1 -1
- package/dist/runner/session.d.mts +1 -1
- package/dist/runner/session.d.ts +1 -1
- package/dist/runner/source-producer.cjs +3 -1
- package/dist/runner/source-producer.cjs.map +1 -1
- package/dist/runner/types.cjs.map +1 -1
- package/dist/runner/types.d.cts +26 -5
- package/dist/runner/types.d.mts +26 -5
- package/dist/runner/types.d.ts +26 -5
- package/dist/sandbox/action.cjs +3 -0
- package/dist/sandbox/action.cjs.map +1 -1
- package/dist/sandbox/commands.cjs +3 -0
- package/dist/sandbox/commands.cjs.map +1 -1
- package/dist/sandbox/compose.cjs +6 -0
- package/dist/sandbox/compose.cjs.map +1 -1
- package/dist/sandbox/docker-profile/registry.cjs +7 -0
- package/dist/sandbox/docker-profile/registry.cjs.map +1 -1
- package/dist/sandbox/docker-profile/runtime.cjs +3 -0
- package/dist/sandbox/docker-profile/runtime.cjs.map +1 -1
- package/dist/sandbox/docker-profile/schema.cjs +7 -0
- package/dist/sandbox/docker-profile/schema.cjs.map +1 -1
- package/dist/sandbox/incus.cjs +7 -0
- package/dist/sandbox/incus.cjs.map +1 -1
- package/dist/sandbox/layer.cjs +12 -0
- package/dist/sandbox/layer.cjs.map +1 -1
- package/dist/shared/adapter-cleanup.cjs +18 -0
- package/dist/shared/adapter-cleanup.cjs.map +1 -0
- package/dist/shared/adapter-cleanup.d.cts +4 -0
- package/dist/shared/adapter-cleanup.d.mts +4 -0
- package/dist/shared/adapter-cleanup.d.ts +4 -0
- package/dist/shared/adapter-cleanup.js +10 -0
- package/dist/shared/adapter-cleanup.js.map +1 -0
- package/dist/shared/attempt-lifecycle.cjs +3 -0
- package/dist/shared/attempt-lifecycle.cjs.map +1 -0
- package/dist/shared/attempt-lifecycle.d.cts +13 -0
- package/dist/shared/attempt-lifecycle.d.mts +13 -0
- package/dist/shared/attempt-lifecycle.d.ts +13 -0
- package/dist/shared/attempt-lifecycle.js +4 -0
- package/dist/shared/attempt-lifecycle.js.map +1 -0
- package/dist/show/contribution.cjs +52 -14
- package/dist/show/contribution.cjs.map +1 -1
- package/dist/show/contribution.d.cts +23 -1
- package/dist/show/contribution.d.mts +23 -1
- package/dist/show/contribution.d.ts +23 -1
- package/dist/show/model.cjs +8 -1
- package/dist/show/model.cjs.map +1 -1
- package/dist/show/model.d.cts +10 -1
- package/dist/show/model.d.mts +10 -1
- package/dist/show/model.d.ts +10 -1
- package/dist/show/render.cjs +287 -49
- package/dist/show/render.cjs.map +1 -1
- package/dist/show/render.d.cts +3 -0
- package/dist/show/render.d.mts +3 -0
- package/dist/show/render.d.ts +3 -0
- package/dist/sources/projector.cjs +3 -0
- package/dist/sources/projector.cjs.map +1 -1
- package/dist/types.cjs.map +1 -1
- package/dist/types.d.cts +1 -1
- package/dist/types.d.mts +1 -1
- package/dist/types.d.ts +1 -1
- package/dist/view/app-dist/.vite/manifest.json +2 -2
- package/dist/view/app-dist/assets/index-CmVoXbwG.js +63 -0
- package/dist/view/app-dist/assets/{quality-cost-scatter-CFB3wnkN.js → quality-cost-scatter-BuLISaH4.js} +1 -1
- package/dist/view/app-dist/index.html +1 -1
- package/dist/view/cli/contribution.cjs +54 -21
- package/dist/view/cli/contribution.cjs.map +1 -1
- package/dist/view/cli/contribution.d.cts +2 -1
- package/dist/view/cli/contribution.d.mts +2 -1
- package/dist/view/cli/contribution.d.ts +2 -1
- package/dist/view/function-runtime.mjs +13231 -11647
- package/dist/view/http-protocol.cjs +3 -0
- package/dist/view/http-protocol.cjs.map +1 -1
- package/dist/view/http-protocol.d.cts +8 -0
- package/dist/view/http-protocol.d.mts +8 -0
- package/dist/view/http-protocol.d.ts +8 -0
- package/dist/view/inspection-host.cjs +17 -1
- package/dist/view/inspection-host.cjs.map +1 -1
- package/dist/view/render.cjs +1 -0
- package/dist/view/render.cjs.map +1 -1
- package/dist/view/render.d.cts +2 -1
- package/dist/view/render.d.mts +2 -1
- package/dist/view/render.d.ts +2 -1
- package/dist/view/revision.cjs.map +1 -1
- package/dist/view/revision.d.cts +12 -0
- package/dist/view/revision.d.mts +12 -0
- package/dist/view/revision.d.ts +12 -0
- package/dist/view/server.cjs +10 -9
- package/dist/view/server.cjs.map +1 -1
- package/docs-site/zh/examples/ai-agent-application.mdx +1 -1
- package/docs-site/zh/explanation/adapter.mdx +2 -2
- package/docs-site/zh/explanation/assert.mdx +60 -24
- package/docs-site/zh/explanation/drive.mdx +1 -1
- package/docs-site/zh/explanation/evals.mdx +31 -29
- package/docs-site/zh/explanation/experiment.mdx +15 -15
- package/docs-site/zh/explanation/judge.mdx +23 -15
- package/docs-site/zh/explanation/overview.mdx +44 -21
- package/docs-site/zh/explanation/record.mdx +28 -13
- package/docs-site/zh/explanation/results-flow.mdx +37 -0
- package/docs-site/zh/explanation/runner.mdx +27 -6
- package/docs-site/zh/index.mdx +22 -115
- package/docs-site/zh/introduction.mdx +89 -61
- package/docs-site/zh/reference/cli.mdx +50 -26
- package/docs-site/zh/reference/define-agent.mdx +37 -5
- package/docs-site/zh/reference/define-eval.mdx +130 -29
- package/docs-site/zh/reference/define-experiment.mdx +251 -0
- package/docs-site/zh/reference/events.mdx +34 -30
- package/docs-site/zh/reference/expect.mdx +77 -2
- package/docs-site/zh/reference/official-adapters.mdx +2 -2
- package/docs-site/zh/troubleshooting/debug-sandbox.mdx +43 -24
- package/docs-site/zh/troubleshooting/debugging.mdx +65 -42
- package/docs-site/zh/troubleshooting/recover-after-kill.mdx +35 -28
- package/docs-site/zh/tutorials/accept-results.mdx +34 -24
- package/docs-site/zh/tutorials/agent-feedback-loop.mdx +74 -43
- package/docs-site/zh/tutorials/agent-onboarding.mdx +27 -20
- package/docs-site/zh/tutorials/authoring.mdx +138 -121
- package/docs-site/zh/tutorials/ci-integration.mdx +46 -22
- package/docs-site/zh/tutorials/compare-runs.mdx +40 -13
- package/docs-site/zh/tutorials/concurrency.mdx +66 -62
- package/docs-site/zh/tutorials/configuration.mdx +17 -17
- package/docs-site/zh/tutorials/connect-otel.mdx +48 -37
- package/docs-site/zh/tutorials/connect-your-agent.mdx +80 -173
- package/docs-site/zh/tutorials/control-run-cost.mdx +63 -12
- package/docs-site/zh/tutorials/criteria-files.mdx +28 -21
- package/docs-site/zh/tutorials/custom-application.mdx +75 -55
- package/docs-site/zh/tutorials/dataset-fanout.mdx +32 -22
- package/docs-site/zh/tutorials/debug-lifecycle-plan.mdx +38 -31
- package/docs-site/zh/tutorials/docker-in-docker.mdx +83 -89
- package/docs-site/zh/tutorials/eval-groups.mdx +70 -89
- package/docs-site/zh/tutorials/evaluation-kinds.mdx +12 -5
- package/docs-site/zh/tutorials/experiment-services.mdx +243 -0
- package/docs-site/zh/tutorials/fixtures.mdx +42 -26
- package/docs-site/zh/tutorials/handle-execution-failures.mdx +41 -19
- package/docs-site/zh/tutorials/install-custom-sandbox-agent.mdx +21 -16
- package/docs-site/zh/tutorials/niceeval-taste-application.mdx +394 -0
- package/docs-site/zh/tutorials/niceeval-taste.mdx +190 -0
- package/docs-site/zh/tutorials/nixos-managed-dind.mdx +51 -39
- package/docs-site/zh/tutorials/plugins.mdx +55 -31
- package/docs-site/zh/tutorials/prebuilt-sandbox.mdx +204 -0
- package/docs-site/zh/tutorials/quickstart.mdx +91 -47
- package/docs-site/zh/tutorials/reporters.mdx +45 -30
- package/docs-site/zh/tutorials/rerun-and-cache.mdx +39 -40
- package/docs-site/zh/tutorials/sandbox-agent.mdx +65 -40
- package/docs-site/zh/tutorials/sandbox-providers.mdx +104 -282
- package/docs-site/zh/tutorials/sandbox-reuse.mdx +65 -97
- package/docs-site/zh/tutorials/select-and-preview.mdx +25 -20
- package/docs-site/zh/tutorials/verify-judge.mdx +49 -26
- package/docs-site/zh/tutorials/viewing-results.mdx +130 -80
- package/docs-site/zh/tutorials/write-experiment.mdx +116 -298
- package/docs-site/zh/tutorials/write-send.mdx +53 -52
- package/node_modules/undici/docs/docs/api/Agent.md +8 -8
- package/node_modules/undici/docs/docs/api/DiagnosticsChannel.md +9 -6
- package/node_modules/undici/docs/docs/api/Errors.md +25 -0
- package/node_modules/undici/docs/docs/api/Interceptors.md +4 -3
- package/node_modules/undici/docs/docs/api/Pool.md +5 -6
- package/node_modules/undici/lib/api/readable.js +1 -2
- package/node_modules/undici/lib/core/request.js +23 -2
- package/node_modules/undici/lib/core/util.js +0 -39
- package/node_modules/undici/lib/dispatcher/client-h1.js +11 -4
- package/node_modules/undici/lib/dispatcher/client-h2.js +52 -8
- package/node_modules/undici/lib/dispatcher/client.js +1 -1
- package/node_modules/undici/lib/dispatcher/dispatcher1-wrapper.js +3 -2
- package/node_modules/undici/lib/dispatcher/env-http-proxy-agent.js +21 -6
- package/node_modules/undici/lib/dispatcher/pool-base.js +63 -2
- package/node_modules/undici/lib/dispatcher/pool.js +90 -7
- package/node_modules/undici/lib/dispatcher/proxy-agent.js +16 -3
- package/node_modules/undici/lib/handler/deduplication-handler.js +24 -11
- package/node_modules/undici/lib/handler/redirect-handler.js +5 -0
- package/node_modules/undici/lib/handler/retry-handler.js +72 -13
- package/node_modules/undici/lib/interceptor/cache.js +4 -1
- package/node_modules/undici/lib/interceptor/decompress.js +236 -69
- package/node_modules/undici/lib/web/cache/cache.js +17 -8
- package/node_modules/undici/lib/web/cache/cachestorage.js +11 -5
- package/node_modules/undici/lib/web/cookies/index.js +19 -7
- package/node_modules/undici/lib/web/cookies/util.js +3 -1
- package/node_modules/undici/lib/web/eventsource/eventsource.js +25 -1
- package/node_modules/undici/lib/web/fetch/body.js +22 -15
- package/node_modules/undici/lib/web/fetch/data-url.js +6 -6
- package/node_modules/undici/lib/web/fetch/formdata.js +30 -36
- package/node_modules/undici/lib/web/fetch/headers.js +34 -31
- package/node_modules/undici/lib/web/fetch/index.js +38 -14
- package/node_modules/undici/lib/web/fetch/request.js +74 -79
- package/node_modules/undici/lib/web/fetch/response.js +43 -44
- package/node_modules/undici/lib/web/fetch/util.js +7 -7
- package/node_modules/undici/lib/web/infra/index.js +16 -1
- package/node_modules/undici/lib/web/webidl/index.js +2 -14
- package/node_modules/undici/lib/web/websocket/events.js +43 -26
- package/node_modules/undici/lib/web/websocket/receiver.js +5 -0
- package/node_modules/undici/lib/web/websocket/stream/websocketerror.js +13 -10
- package/node_modules/undici/lib/web/websocket/websocket.js +44 -41
- package/node_modules/undici/package.json +1 -1
- package/node_modules/undici/types/diagnostics-channel.d.ts +2 -1
- package/node_modules/undici/types/interceptors.d.ts +1 -1
- package/node_modules/undici/types/webidl.d.ts +13 -7
- package/package.json +4 -4
- package/dist/view/app-dist/assets/index-DFzw1L97.js +0 -63
- package/docs-site/zh/tutorials/experiments.mdx +0 -101
package/dist/runner/types.d.cts
CHANGED
|
@@ -1,3 +1,4 @@
|
|
|
1
|
+
import type { ModelSlotSelection, ResolvedModelSlots } from "../model-slots.cjs";
|
|
1
2
|
import type { AdapterFlagsParser, AdapterFlagsOutput, AdapterFlagsValue } from "../adapter-flags.cjs";
|
|
2
3
|
import type { Effect } from "effect";
|
|
3
4
|
import type { ExperimentFlags, FlagValue, JsonValue, LocalizedText, ScopedFeedback, SourceArtifact, Verdict } from "../shared/types.cjs";
|
|
@@ -22,6 +23,7 @@ import type { PluginInstance, PluginOnUnavailable } from "../plugin/contracts.cj
|
|
|
22
23
|
* 这里不复制(见 docs/feature/record/architecture.md「run.json」)。
|
|
23
24
|
*/
|
|
24
25
|
export interface ExperimentRunInfo {
|
|
26
|
+
readonly models?: ResolvedModelSlots;
|
|
25
27
|
description?: string;
|
|
26
28
|
reasoningEffort?: string;
|
|
27
29
|
flags?: globalThis.Record<string, JsonValue>;
|
|
@@ -814,11 +816,13 @@ export interface SandboxCacheConfig {
|
|
|
814
816
|
export declare function resolveSandboxSetupCache(cli: SandboxSetupCache | undefined, experiment: SandboxCacheConfig | undefined, config: SandboxCacheConfig | undefined): SandboxSetupCache;
|
|
815
817
|
/** Experiment 作者自行选择的字段;不包含路径 id 与 factory 品牌。 */
|
|
816
818
|
export interface ExperimentAuthorFields {
|
|
819
|
+
/** 为自定义 Adapter 的各个具名模型槽选择模型;不能与单模型简写同时使用。 */
|
|
820
|
+
models?: Readonly<Record<string, ModelSlotSelection>>;
|
|
817
821
|
/** 一句话描述,展示在 view / CLI 里;纯说明,不影响调度或打分。 */
|
|
818
822
|
description?: string;
|
|
819
|
-
/**
|
|
823
|
+
/** 对话 Adapter 的简写;`agent` 与 `adapter` 必须且只能提供一个。 */
|
|
820
824
|
agent?: Agent;
|
|
821
|
-
/**
|
|
825
|
+
/** 本实验使用的 Adapter 实现。 */
|
|
822
826
|
adapter?: Adapter;
|
|
823
827
|
/** 单个模型(agent 留空时实验决定);省略=用 agent 原生默认。跨模型对比写多个实验文件,别用数组。 */
|
|
824
828
|
model?: string;
|
|
@@ -863,9 +867,9 @@ export interface ExperimentAuthorFields {
|
|
|
863
867
|
* 每个配对恰好一方提供 template-bearing layer。
|
|
864
868
|
*/
|
|
865
869
|
sandbox?: SandboxLayer;
|
|
866
|
-
/** Host
|
|
870
|
+
/** Host 的执行缓存策略;与 Sandbox 声明层不同,此字段不参与身份计算。 */
|
|
867
871
|
sandboxCache?: SandboxCacheConfig;
|
|
868
|
-
/**
|
|
872
|
+
/** 本实验显式声明的 Plugin 实例列表,由 `defineExperiment()` 规范化。 */
|
|
869
873
|
plugins?: readonly PluginInstance<"experiment">[];
|
|
870
874
|
/** 同一 Run 内复用沙箱;这种运行与历史携带双向隔离。 */
|
|
871
875
|
sandboxReuse?: boolean;
|
|
@@ -936,11 +940,27 @@ type ExperimentAdapterSelection<A extends Adapter> = {
|
|
|
936
940
|
readonly agent: A & Agent;
|
|
937
941
|
readonly adapter?: never;
|
|
938
942
|
};
|
|
939
|
-
|
|
943
|
+
type ExperimentModelFields<A extends Adapter> = A extends {
|
|
944
|
+
readonly kind: "custom";
|
|
945
|
+
} ? ({
|
|
946
|
+
readonly models: Readonly<Record<string, ModelSlotSelection>>;
|
|
947
|
+
readonly model?: never;
|
|
948
|
+
readonly reasoningEffort?: never;
|
|
949
|
+
} | {
|
|
950
|
+
readonly models?: never;
|
|
951
|
+
readonly model?: string;
|
|
952
|
+
readonly reasoningEffort?: string;
|
|
953
|
+
}) : {
|
|
954
|
+
readonly models?: never;
|
|
955
|
+
readonly model?: string;
|
|
956
|
+
readonly reasoningEffort?: string;
|
|
957
|
+
};
|
|
958
|
+
export type ExperimentInput<A extends Adapter = Adapter> = Omit<ExperimentAuthorFields, "agent" | "adapter" | "flags" | "models" | "model" | "reasoningEffort"> & ExperimentModelFields<A> & ExperimentAdapterSelection<A> & ExperimentFlagsFields<NoInfer<A>> & {
|
|
940
959
|
id?: ExperimentIdComesFromFilePath;
|
|
941
960
|
};
|
|
942
961
|
/** Factory 完成默认归一后的 Experiment 字段;无默认语义的 Hook 仍保持作者声明。 */
|
|
943
962
|
export interface ExperimentDefinition<A extends Adapter = Adapter> {
|
|
963
|
+
readonly models?: ResolvedModelSlots;
|
|
944
964
|
readonly description?: string;
|
|
945
965
|
readonly adapter: A;
|
|
946
966
|
/** Present when the selected Adapter is an Agent, regardless of the input shorthand. */
|
|
@@ -1095,6 +1115,7 @@ export declare function runWho(run: {
|
|
|
1095
1115
|
}): string;
|
|
1096
1116
|
/** 一个 (agent, model, flags) 的运行配置 —— 由 CLI / 实验展开。 */
|
|
1097
1117
|
export interface AdapterRun {
|
|
1118
|
+
readonly models?: ResolvedModelSlots;
|
|
1098
1119
|
readonly adapter: Adapter;
|
|
1099
1120
|
/** Present exactly when `adapter` is an Agent implementation. */
|
|
1100
1121
|
readonly agent?: Agent;
|
package/dist/runner/types.d.mts
CHANGED
|
@@ -1,3 +1,4 @@
|
|
|
1
|
+
import type { ModelSlotSelection, ResolvedModelSlots } from "../model-slots.mjs";
|
|
1
2
|
import type { AdapterFlagsParser, AdapterFlagsOutput, AdapterFlagsValue } from "../adapter-flags.mjs";
|
|
2
3
|
import type { Effect } from "effect";
|
|
3
4
|
import type { ExperimentFlags, FlagValue, JsonValue, LocalizedText, ScopedFeedback, SourceArtifact, Verdict } from "../shared/types.mjs";
|
|
@@ -22,6 +23,7 @@ import type { PluginInstance, PluginOnUnavailable } from "../plugin/contracts.mj
|
|
|
22
23
|
* 这里不复制(见 docs/feature/record/architecture.md「run.json」)。
|
|
23
24
|
*/
|
|
24
25
|
export interface ExperimentRunInfo {
|
|
26
|
+
readonly models?: ResolvedModelSlots;
|
|
25
27
|
description?: string;
|
|
26
28
|
reasoningEffort?: string;
|
|
27
29
|
flags?: globalThis.Record<string, JsonValue>;
|
|
@@ -814,11 +816,13 @@ export interface SandboxCacheConfig {
|
|
|
814
816
|
export declare function resolveSandboxSetupCache(cli: SandboxSetupCache | undefined, experiment: SandboxCacheConfig | undefined, config: SandboxCacheConfig | undefined): SandboxSetupCache;
|
|
815
817
|
/** Experiment 作者自行选择的字段;不包含路径 id 与 factory 品牌。 */
|
|
816
818
|
export interface ExperimentAuthorFields {
|
|
819
|
+
/** 为自定义 Adapter 的各个具名模型槽选择模型;不能与单模型简写同时使用。 */
|
|
820
|
+
models?: Readonly<Record<string, ModelSlotSelection>>;
|
|
817
821
|
/** 一句话描述,展示在 view / CLI 里;纯说明,不影响调度或打分。 */
|
|
818
822
|
description?: string;
|
|
819
|
-
/**
|
|
823
|
+
/** 对话 Adapter 的简写;`agent` 与 `adapter` 必须且只能提供一个。 */
|
|
820
824
|
agent?: Agent;
|
|
821
|
-
/**
|
|
825
|
+
/** 本实验使用的 Adapter 实现。 */
|
|
822
826
|
adapter?: Adapter;
|
|
823
827
|
/** 单个模型(agent 留空时实验决定);省略=用 agent 原生默认。跨模型对比写多个实验文件,别用数组。 */
|
|
824
828
|
model?: string;
|
|
@@ -863,9 +867,9 @@ export interface ExperimentAuthorFields {
|
|
|
863
867
|
* 每个配对恰好一方提供 template-bearing layer。
|
|
864
868
|
*/
|
|
865
869
|
sandbox?: SandboxLayer;
|
|
866
|
-
/** Host
|
|
870
|
+
/** Host 的执行缓存策略;与 Sandbox 声明层不同,此字段不参与身份计算。 */
|
|
867
871
|
sandboxCache?: SandboxCacheConfig;
|
|
868
|
-
/**
|
|
872
|
+
/** 本实验显式声明的 Plugin 实例列表,由 `defineExperiment()` 规范化。 */
|
|
869
873
|
plugins?: readonly PluginInstance<"experiment">[];
|
|
870
874
|
/** 同一 Run 内复用沙箱;这种运行与历史携带双向隔离。 */
|
|
871
875
|
sandboxReuse?: boolean;
|
|
@@ -936,11 +940,27 @@ type ExperimentAdapterSelection<A extends Adapter> = {
|
|
|
936
940
|
readonly agent: A & Agent;
|
|
937
941
|
readonly adapter?: never;
|
|
938
942
|
};
|
|
939
|
-
|
|
943
|
+
type ExperimentModelFields<A extends Adapter> = A extends {
|
|
944
|
+
readonly kind: "custom";
|
|
945
|
+
} ? ({
|
|
946
|
+
readonly models: Readonly<Record<string, ModelSlotSelection>>;
|
|
947
|
+
readonly model?: never;
|
|
948
|
+
readonly reasoningEffort?: never;
|
|
949
|
+
} | {
|
|
950
|
+
readonly models?: never;
|
|
951
|
+
readonly model?: string;
|
|
952
|
+
readonly reasoningEffort?: string;
|
|
953
|
+
}) : {
|
|
954
|
+
readonly models?: never;
|
|
955
|
+
readonly model?: string;
|
|
956
|
+
readonly reasoningEffort?: string;
|
|
957
|
+
};
|
|
958
|
+
export type ExperimentInput<A extends Adapter = Adapter> = Omit<ExperimentAuthorFields, "agent" | "adapter" | "flags" | "models" | "model" | "reasoningEffort"> & ExperimentModelFields<A> & ExperimentAdapterSelection<A> & ExperimentFlagsFields<NoInfer<A>> & {
|
|
940
959
|
id?: ExperimentIdComesFromFilePath;
|
|
941
960
|
};
|
|
942
961
|
/** Factory 完成默认归一后的 Experiment 字段;无默认语义的 Hook 仍保持作者声明。 */
|
|
943
962
|
export interface ExperimentDefinition<A extends Adapter = Adapter> {
|
|
963
|
+
readonly models?: ResolvedModelSlots;
|
|
944
964
|
readonly description?: string;
|
|
945
965
|
readonly adapter: A;
|
|
946
966
|
/** Present when the selected Adapter is an Agent, regardless of the input shorthand. */
|
|
@@ -1095,6 +1115,7 @@ export declare function runWho(run: {
|
|
|
1095
1115
|
}): string;
|
|
1096
1116
|
/** 一个 (agent, model, flags) 的运行配置 —— 由 CLI / 实验展开。 */
|
|
1097
1117
|
export interface AdapterRun {
|
|
1118
|
+
readonly models?: ResolvedModelSlots;
|
|
1098
1119
|
readonly adapter: Adapter;
|
|
1099
1120
|
/** Present exactly when `adapter` is an Agent implementation. */
|
|
1100
1121
|
readonly agent?: Agent;
|
package/dist/runner/types.d.ts
CHANGED
|
@@ -1,3 +1,4 @@
|
|
|
1
|
+
import type { ModelSlotSelection, ResolvedModelSlots } from "../model-slots.js";
|
|
1
2
|
import type { AdapterFlagsParser, AdapterFlagsOutput, AdapterFlagsValue } from "../adapter-flags.js";
|
|
2
3
|
import type { Effect } from "effect";
|
|
3
4
|
import type { ExperimentFlags, FlagValue, JsonValue, LocalizedText, ScopedFeedback, SourceArtifact, Verdict } from "../shared/types.js";
|
|
@@ -22,6 +23,7 @@ import type { PluginInstance, PluginOnUnavailable } from "../plugin/contracts.js
|
|
|
22
23
|
* 这里不复制(见 docs/feature/record/architecture.md「run.json」)。
|
|
23
24
|
*/
|
|
24
25
|
export interface ExperimentRunInfo {
|
|
26
|
+
readonly models?: ResolvedModelSlots;
|
|
25
27
|
description?: string;
|
|
26
28
|
reasoningEffort?: string;
|
|
27
29
|
flags?: globalThis.Record<string, JsonValue>;
|
|
@@ -814,11 +816,13 @@ export interface SandboxCacheConfig {
|
|
|
814
816
|
export declare function resolveSandboxSetupCache(cli: SandboxSetupCache | undefined, experiment: SandboxCacheConfig | undefined, config: SandboxCacheConfig | undefined): SandboxSetupCache;
|
|
815
817
|
/** Experiment 作者自行选择的字段;不包含路径 id 与 factory 品牌。 */
|
|
816
818
|
export interface ExperimentAuthorFields {
|
|
819
|
+
/** 为自定义 Adapter 的各个具名模型槽选择模型;不能与单模型简写同时使用。 */
|
|
820
|
+
models?: Readonly<Record<string, ModelSlotSelection>>;
|
|
817
821
|
/** 一句话描述,展示在 view / CLI 里;纯说明,不影响调度或打分。 */
|
|
818
822
|
description?: string;
|
|
819
|
-
/**
|
|
823
|
+
/** 对话 Adapter 的简写;`agent` 与 `adapter` 必须且只能提供一个。 */
|
|
820
824
|
agent?: Agent;
|
|
821
|
-
/**
|
|
825
|
+
/** 本实验使用的 Adapter 实现。 */
|
|
822
826
|
adapter?: Adapter;
|
|
823
827
|
/** 单个模型(agent 留空时实验决定);省略=用 agent 原生默认。跨模型对比写多个实验文件,别用数组。 */
|
|
824
828
|
model?: string;
|
|
@@ -863,9 +867,9 @@ export interface ExperimentAuthorFields {
|
|
|
863
867
|
* 每个配对恰好一方提供 template-bearing layer。
|
|
864
868
|
*/
|
|
865
869
|
sandbox?: SandboxLayer;
|
|
866
|
-
/** Host
|
|
870
|
+
/** Host 的执行缓存策略;与 Sandbox 声明层不同,此字段不参与身份计算。 */
|
|
867
871
|
sandboxCache?: SandboxCacheConfig;
|
|
868
|
-
/**
|
|
872
|
+
/** 本实验显式声明的 Plugin 实例列表,由 `defineExperiment()` 规范化。 */
|
|
869
873
|
plugins?: readonly PluginInstance<"experiment">[];
|
|
870
874
|
/** 同一 Run 内复用沙箱;这种运行与历史携带双向隔离。 */
|
|
871
875
|
sandboxReuse?: boolean;
|
|
@@ -936,11 +940,27 @@ type ExperimentAdapterSelection<A extends Adapter> = {
|
|
|
936
940
|
readonly agent: A & Agent;
|
|
937
941
|
readonly adapter?: never;
|
|
938
942
|
};
|
|
939
|
-
|
|
943
|
+
type ExperimentModelFields<A extends Adapter> = A extends {
|
|
944
|
+
readonly kind: "custom";
|
|
945
|
+
} ? ({
|
|
946
|
+
readonly models: Readonly<Record<string, ModelSlotSelection>>;
|
|
947
|
+
readonly model?: never;
|
|
948
|
+
readonly reasoningEffort?: never;
|
|
949
|
+
} | {
|
|
950
|
+
readonly models?: never;
|
|
951
|
+
readonly model?: string;
|
|
952
|
+
readonly reasoningEffort?: string;
|
|
953
|
+
}) : {
|
|
954
|
+
readonly models?: never;
|
|
955
|
+
readonly model?: string;
|
|
956
|
+
readonly reasoningEffort?: string;
|
|
957
|
+
};
|
|
958
|
+
export type ExperimentInput<A extends Adapter = Adapter> = Omit<ExperimentAuthorFields, "agent" | "adapter" | "flags" | "models" | "model" | "reasoningEffort"> & ExperimentModelFields<A> & ExperimentAdapterSelection<A> & ExperimentFlagsFields<NoInfer<A>> & {
|
|
940
959
|
id?: ExperimentIdComesFromFilePath;
|
|
941
960
|
};
|
|
942
961
|
/** Factory 完成默认归一后的 Experiment 字段;无默认语义的 Hook 仍保持作者声明。 */
|
|
943
962
|
export interface ExperimentDefinition<A extends Adapter = Adapter> {
|
|
963
|
+
readonly models?: ResolvedModelSlots;
|
|
944
964
|
readonly description?: string;
|
|
945
965
|
readonly adapter: A;
|
|
946
966
|
/** Present when the selected Adapter is an Agent, regardless of the input shorthand. */
|
|
@@ -1095,6 +1115,7 @@ export declare function runWho(run: {
|
|
|
1095
1115
|
}): string;
|
|
1096
1116
|
/** 一个 (agent, model, flags) 的运行配置 —— 由 CLI / 实验展开。 */
|
|
1097
1117
|
export interface AdapterRun {
|
|
1118
|
+
readonly models?: ResolvedModelSlots;
|
|
1098
1119
|
readonly adapter: Adapter;
|
|
1099
1120
|
/** Present exactly when `adapter` is an Agent implementation. */
|
|
1100
1121
|
readonly agent?: Agent;
|
package/dist/sandbox/action.cjs
CHANGED
|
@@ -547,6 +547,9 @@ function makeAfterAction(data) {
|
|
|
547
547
|
SANDBOX_AFTER_ACTION_DATA.set(action, data);
|
|
548
548
|
return Object.freeze(action);
|
|
549
549
|
}
|
|
550
|
+
// @concord-code ne-sandbox-define-action
|
|
551
|
+
// @concord-implements docs/feature/sandbox/prepare-commands.md
|
|
552
|
+
// @concord-implements docs/feature/sandbox/architecture.md
|
|
550
553
|
function defineSandboxAction(definition) {
|
|
551
554
|
assertRecord(definition, "defineSandboxAction definition");
|
|
552
555
|
assertOnlyKeys(definition, new Set(["id", "input", "cache", "steps"]), "defineSandboxAction definition");
|