@bjoernrochel/sevro 0.1.0-rc.8 → 1.1.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CONTRIBUTING.md +52 -6
- package/README.md +12 -7
- package/dist/examples/basic/graded.json +13 -0
- package/dist/examples/basic/host.js +9 -0
- package/dist/examples/basic/host.js.map +1 -0
- package/dist/examples/basic/prompt-only.json +7 -0
- package/dist/examples/recipes/candidate.js +23 -0
- package/dist/examples/recipes/candidate.js.map +1 -0
- package/dist/examples/recipes/cases/coding-change.json +21 -0
- package/dist/examples/recipes/cases/host-tools.json +21 -0
- package/dist/examples/recipes/cases/mise.json +21 -0
- package/dist/examples/recipes/cases/node.json +21 -0
- package/dist/examples/recipes/cases/python-uv.json +21 -0
- package/dist/examples/recipes/cases/semantic.json +17 -0
- package/dist/examples/recipes/coding-change/check.js +16 -0
- package/dist/examples/recipes/coding-change/check.js.map +1 -0
- package/dist/examples/recipes/executable.js +20 -0
- package/dist/examples/recipes/executable.js.map +1 -0
- package/dist/examples/recipes/extension-artifacts.js +56 -0
- package/dist/examples/recipes/extension-artifacts.js.map +1 -0
- package/dist/examples/recipes/extension-methods.js +167 -0
- package/dist/examples/recipes/extension-methods.js.map +1 -0
- package/dist/examples/recipes/extension.js +28 -0
- package/dist/examples/recipes/extension.js.map +1 -0
- package/dist/examples/recipes/modules.js +9 -0
- package/dist/examples/recipes/modules.js.map +1 -0
- package/dist/examples/recipes/repository-skill/.agents/skills/startup-guide/SKILL.md +7 -0
- package/dist/examples/recipes/repository-skill/.claude/skills/startup-guide/SKILL.md +7 -0
- package/dist/examples/recipes/repository-skill/README.md +3 -0
- package/dist/examples/recipes/run-inputs.js +144 -0
- package/dist/examples/recipes/run-inputs.js.map +1 -0
- package/dist/examples/recipes/run-options.js +103 -0
- package/dist/examples/recipes/run-options.js.map +1 -0
- package/dist/examples/recipes/run.js +62 -0
- package/dist/examples/recipes/run.js.map +1 -0
- package/dist/examples/recipes/runtime/empty.json +1 -0
- package/dist/examples/recipes/runtime/goals.json +1 -0
- package/dist/examples/recipes/runtime/mise.json +6 -0
- package/dist/examples/recipes/runtime/path.json +1 -0
- package/dist/examples/recipes/runtime/plugins.json +4 -0
- package/dist/examples/recipes/runtime/python-uv.json +10 -0
- package/dist/examples/recipes/semantic-judge.js +27 -0
- package/dist/examples/recipes/semantic-judge.js.map +1 -0
- package/dist/examples/templates/coding/adapter.js +14 -0
- package/dist/examples/templates/coding/adapter.js.map +1 -0
- package/dist/examples/templates/coding/case.json +15 -0
- package/dist/examples/templates/repository-skill/.agents/skills/startup-guide/SKILL.md +6 -0
- package/dist/examples/templates/repository-skill/extension-data.js +61 -0
- package/dist/examples/templates/repository-skill/extension-data.js.map +1 -0
- package/dist/examples/templates/repository-skill/extension.js +29 -0
- package/dist/examples/templates/repository-skill/extension.js.map +1 -0
- package/dist/examples/templates/response/adapter.js +13 -0
- package/dist/examples/templates/response/adapter.js.map +1 -0
- package/dist/examples/templates/response/case.json +13 -0
- package/dist/src/advisory.js +86 -0
- package/dist/src/advisory.js.map +1 -0
- package/dist/src/candidate-artifacts.js +139 -0
- package/dist/src/candidate-artifacts.js.map +1 -0
- package/dist/src/candidate-codex-state.js +110 -0
- package/dist/src/candidate-codex-state.js.map +1 -0
- package/dist/src/case-validation.js +92 -0
- package/dist/src/case-validation.js.map +1 -0
- package/dist/src/cli/cancellation.js +21 -0
- package/dist/src/cli/cancellation.js.map +1 -0
- package/dist/src/cli/command.js +60 -0
- package/dist/src/cli/command.js.map +1 -0
- package/dist/src/cli/completion.js +14 -0
- package/dist/src/cli/completion.js.map +1 -0
- package/dist/src/cli/demo.js +45 -0
- package/dist/src/cli/demo.js.map +1 -0
- package/dist/src/cli/evaluation.js +40 -0
- package/dist/src/cli/evaluation.js.map +1 -0
- package/dist/src/cli/extension.js +54 -0
- package/dist/src/cli/extension.js.map +1 -0
- package/dist/src/cli/help.js +190 -0
- package/dist/src/cli/help.js.map +1 -0
- package/dist/src/cli/hosts.js +33 -0
- package/dist/src/cli/hosts.js.map +1 -0
- package/dist/src/cli/init-settings.js +29 -0
- package/dist/src/cli/init-settings.js.map +1 -0
- package/dist/src/cli/init.js +82 -0
- package/dist/src/cli/init.js.map +1 -0
- package/dist/src/cli/invocation.js +431 -0
- package/dist/src/cli/invocation.js.map +1 -0
- package/dist/src/cli/loading.js +110 -0
- package/dist/src/cli/loading.js.map +1 -0
- package/dist/src/cli/preparation.js +88 -0
- package/dist/src/cli/preparation.js.map +1 -0
- package/dist/src/cli/presentation.js +61 -0
- package/dist/src/cli/presentation.js.map +1 -0
- package/dist/src/cli/progress.js +5 -0
- package/dist/src/cli/progress.js.map +1 -0
- package/dist/src/cli/run-manifest.js +108 -0
- package/dist/src/cli/run-manifest.js.map +1 -0
- package/dist/src/cli/run-options.js +367 -0
- package/dist/src/cli/run-options.js.map +1 -0
- package/dist/src/cli/run-overrides.js +51 -0
- package/dist/src/cli/run-overrides.js.map +1 -0
- package/dist/src/cli/settings.js +127 -0
- package/dist/src/cli/settings.js.map +1 -0
- package/dist/src/cli/summary.js +28 -0
- package/dist/src/cli/summary.js.map +1 -0
- package/dist/src/cli.js +20 -0
- package/dist/src/cli.js.map +1 -0
- package/dist/src/contract-types.js +2 -0
- package/dist/src/contract-types.js.map +1 -0
- package/dist/src/engine.js +4 -0
- package/dist/src/engine.js.map +1 -0
- package/dist/src/evaluation/advisory.js +85 -0
- package/dist/src/evaluation/advisory.js.map +1 -0
- package/dist/src/evaluation/check-plan.js +292 -0
- package/dist/src/evaluation/check-plan.js.map +1 -0
- package/dist/src/evaluation/collection.js +61 -0
- package/dist/src/evaluation/collection.js.map +1 -0
- package/dist/src/evaluation/context.js +2 -0
- package/dist/src/evaluation/context.js.map +1 -0
- package/dist/src/evaluation/grading-host.js +28 -0
- package/dist/src/evaluation/grading-host.js.map +1 -0
- package/dist/src/evaluation/grading.js +154 -0
- package/dist/src/evaluation/grading.js.map +1 -0
- package/dist/src/evaluation/invocation-plan.js +312 -0
- package/dist/src/evaluation/invocation-plan.js.map +1 -0
- package/dist/src/evaluation/plan.js +376 -0
- package/dist/src/evaluation/plan.js.map +1 -0
- package/dist/src/evaluation/progress.js +14 -0
- package/dist/src/evaluation/progress.js.map +1 -0
- package/dist/src/evaluation/resources.js +159 -0
- package/dist/src/evaluation/resources.js.map +1 -0
- package/dist/src/evaluation/run.js +148 -0
- package/dist/src/evaluation/run.js.map +1 -0
- package/dist/src/evaluation/scheduler.js +27 -0
- package/dist/src/evaluation/scheduler.js.map +1 -0
- package/dist/src/evaluation/semantic.js +183 -0
- package/dist/src/evaluation/semantic.js.map +1 -0
- package/dist/src/evaluation/trial.js +593 -0
- package/dist/src/evaluation/trial.js.map +1 -0
- package/dist/src/evaluation/types.js +20 -0
- package/dist/src/evaluation/types.js.map +1 -0
- package/dist/src/evidence/host-results.js +67 -0
- package/dist/src/evidence/host-results.js.map +1 -0
- package/dist/src/evidence/retained-artifacts.js +17 -0
- package/dist/src/evidence/retained-artifacts.js.map +1 -0
- package/dist/src/evidence/trial-writer.js +131 -0
- package/dist/src/evidence/trial-writer.js.map +1 -0
- package/dist/src/evidence/usage.js +14 -0
- package/dist/src/evidence/usage.js.map +1 -0
- package/dist/src/executable.js +20 -0
- package/dist/src/executable.js.map +1 -0
- package/dist/src/extension-client.js +212 -0
- package/dist/src/extension-client.js.map +1 -0
- package/dist/src/extension-session.js +236 -0
- package/dist/src/extension-session.js.map +1 -0
- package/dist/src/fixtures/advisory.js +206 -0
- package/dist/src/fixtures/advisory.js.map +1 -0
- package/dist/src/fixtures/bin.js +41 -0
- package/dist/src/fixtures/bin.js.map +1 -0
- package/dist/src/fixtures/cleanup.js +43 -0
- package/dist/src/fixtures/cleanup.js.map +1 -0
- package/dist/src/fixtures/create.js +94 -0
- package/dist/src/fixtures/create.js.map +1 -0
- package/dist/src/fixtures/generated.js +181 -0
- package/dist/src/fixtures/generated.js.map +1 -0
- package/dist/src/fixtures/git-hooks.js +16 -0
- package/dist/src/fixtures/git-hooks.js.map +1 -0
- package/dist/src/fixtures/preparation.js +201 -0
- package/dist/src/fixtures/preparation.js.map +1 -0
- package/dist/src/fixtures/repository.js +255 -0
- package/dist/src/fixtures/repository.js.map +1 -0
- package/dist/src/fixtures/setup.js +175 -0
- package/dist/src/fixtures/setup.js.map +1 -0
- package/dist/src/fixtures/tools.js +57 -0
- package/dist/src/fixtures/tools.js.map +1 -0
- package/dist/src/fixtures/workspace-state.js +155 -0
- package/dist/src/fixtures/workspace-state.js.map +1 -0
- package/dist/src/generated/cli-result.cjs +2 -0
- package/dist/src/generated/extension.cjs +2 -0
- package/dist/src/generated/report.cjs +2 -0
- package/dist/src/generated/run-evidence.cjs +2 -0
- package/dist/src/generated/runtime.cjs +2 -0
- package/dist/src/git.js +72 -0
- package/dist/src/git.js.map +1 -0
- package/dist/src/graders/artifact.js +68 -0
- package/dist/src/graders/artifact.js.map +1 -0
- package/dist/src/graders/git-head.js +112 -0
- package/dist/src/graders/git-head.js.map +1 -0
- package/dist/src/graders/output.js +345 -0
- package/dist/src/graders/output.js.map +1 -0
- package/dist/src/graders/semantic.js +132 -0
- package/dist/src/graders/semantic.js.map +1 -0
- package/dist/src/graders/shell.js +321 -0
- package/dist/src/graders/shell.js.map +1 -0
- package/dist/src/host-diagnostic-redaction.js +52 -0
- package/dist/src/host-diagnostic-redaction.js.map +1 -0
- package/dist/src/host-diagnostics.js +248 -0
- package/dist/src/host-diagnostics.js.map +1 -0
- package/dist/src/hosts/claude-continuation.js +89 -0
- package/dist/src/hosts/claude-continuation.js.map +1 -0
- package/dist/src/hosts/claude-credential.js +121 -0
- package/dist/src/hosts/claude-credential.js.map +1 -0
- package/dist/src/hosts/claude-events.js +136 -0
- package/dist/src/hosts/claude-events.js.map +1 -0
- package/dist/src/hosts/claude-keychain.js +100 -0
- package/dist/src/hosts/claude-keychain.js.map +1 -0
- package/dist/src/hosts/claude-login-recovery.js +67 -0
- package/dist/src/hosts/claude-login-recovery.js.map +1 -0
- package/dist/src/hosts/claude-login-refresh.js +146 -0
- package/dist/src/hosts/claude-login-refresh.js.map +1 -0
- package/dist/src/hosts/claude-login-store.js +124 -0
- package/dist/src/hosts/claude-login-store.js.map +1 -0
- package/dist/src/hosts/claude-login.js +74 -0
- package/dist/src/hosts/claude-login.js.map +1 -0
- package/dist/src/hosts/claude-native-goal.js +138 -0
- package/dist/src/hosts/claude-native-goal.js.map +1 -0
- package/dist/src/hosts/claude-process.js +10 -0
- package/dist/src/hosts/claude-process.js.map +1 -0
- package/dist/src/hosts/claude-refresh-lock.js +76 -0
- package/dist/src/hosts/claude-refresh-lock.js.map +1 -0
- package/dist/src/hosts/claude-repository-invocation.js +229 -0
- package/dist/src/hosts/claude-repository-invocation.js.map +1 -0
- package/dist/src/hosts/claude-settings.js +156 -0
- package/dist/src/hosts/claude-settings.js.map +1 -0
- package/dist/src/hosts/claude-skill-activations.js +131 -0
- package/dist/src/hosts/claude-skill-activations.js.map +1 -0
- package/dist/src/hosts/claude-state.js +237 -0
- package/dist/src/hosts/claude-state.js.map +1 -0
- package/dist/src/hosts/claude-tool-calls.js +150 -0
- package/dist/src/hosts/claude-tool-calls.js.map +1 -0
- package/dist/src/hosts/claude.js +373 -0
- package/dist/src/hosts/claude.js.map +1 -0
- package/dist/src/hosts/codex-app-server.js +764 -0
- package/dist/src/hosts/codex-app-server.js.map +1 -0
- package/dist/src/hosts/codex-configuration.js +91 -0
- package/dist/src/hosts/codex-configuration.js.map +1 -0
- package/dist/src/hosts/codex-events.js +104 -0
- package/dist/src/hosts/codex-events.js.map +1 -0
- package/dist/src/hosts/codex-exec-controls.js +434 -0
- package/dist/src/hosts/codex-exec-controls.js.map +1 -0
- package/dist/src/hosts/codex-invocation.js +54 -0
- package/dist/src/hosts/codex-invocation.js.map +1 -0
- package/dist/src/hosts/codex-marketplace.js +185 -0
- package/dist/src/hosts/codex-marketplace.js.map +1 -0
- package/dist/src/hosts/codex-native-calls.js +516 -0
- package/dist/src/hosts/codex-native-calls.js.map +1 -0
- package/dist/src/hosts/codex-process.js +10 -0
- package/dist/src/hosts/codex-process.js.map +1 -0
- package/dist/src/hosts/codex-profile.js +206 -0
- package/dist/src/hosts/codex-profile.js.map +1 -0
- package/dist/src/hosts/codex-session-files.js +89 -0
- package/dist/src/hosts/codex-session-files.js.map +1 -0
- package/dist/src/hosts/codex-state.js +310 -0
- package/dist/src/hosts/codex-state.js.map +1 -0
- package/dist/src/hosts/codex.js +527 -0
- package/dist/src/hosts/codex.js.map +1 -0
- package/dist/src/hosts/isolation-common.js +43 -0
- package/dist/src/hosts/isolation-common.js.map +1 -0
- package/dist/src/hosts/isolation-roots.js +64 -0
- package/dist/src/hosts/isolation-roots.js.map +1 -0
- package/dist/src/hosts/isolation.js +12 -0
- package/dist/src/hosts/isolation.js.map +1 -0
- package/dist/src/hosts/linux-sandbox.js +127 -0
- package/dist/src/hosts/linux-sandbox.js.map +1 -0
- package/dist/src/hosts/mac-sandbox.js +137 -0
- package/dist/src/hosts/mac-sandbox.js.map +1 -0
- package/dist/src/hosts/native-controls.js +151 -0
- package/dist/src/hosts/native-controls.js.map +1 -0
- package/dist/src/hosts/native-request.js +33 -0
- package/dist/src/hosts/native-request.js.map +1 -0
- package/dist/src/hosts/process-lifecycle.js +167 -0
- package/dist/src/hosts/process-lifecycle.js.map +1 -0
- package/dist/src/hosts/protected-worktrees.js +81 -0
- package/dist/src/hosts/protected-worktrees.js.map +1 -0
- package/dist/src/hosts/skill-hook-catalog.js +100 -0
- package/dist/src/hosts/skill-hook-catalog.js.map +1 -0
- package/dist/src/hosts/skill-hook-handler.js +228 -0
- package/dist/src/hosts/skill-hook-handler.js.map +1 -0
- package/dist/src/hosts/skill-hooks.js +423 -0
- package/dist/src/hosts/skill-hooks.js.map +1 -0
- package/dist/src/hosts/workspace-fingerprint.js +72 -0
- package/dist/src/hosts/workspace-fingerprint.js.map +1 -0
- package/dist/src/identity.js +167 -0
- package/dist/src/identity.js.map +1 -0
- package/dist/src/instrumentation.js +83 -0
- package/dist/src/instrumentation.js.map +1 -0
- package/dist/src/native-transcript-state.js +85 -0
- package/dist/src/native-transcript-state.js.map +1 -0
- package/dist/src/native-transcript-view.js +65 -0
- package/dist/src/native-transcript-view.js.map +1 -0
- package/dist/src/native-transcripts.js +208 -0
- package/dist/src/native-transcripts.js.map +1 -0
- package/dist/src/paths.js +13 -0
- package/dist/src/paths.js.map +1 -0
- package/dist/src/process-lifecycle.js +101 -0
- package/dist/src/process-lifecycle.js.map +1 -0
- package/dist/src/provenance.js +263 -0
- package/dist/src/provenance.js.map +1 -0
- package/dist/src/prune.js +228 -0
- package/dist/src/prune.js.map +1 -0
- package/dist/src/report.js +276 -0
- package/dist/src/report.js.map +1 -0
- package/dist/src/resolved-case.js +29 -0
- package/dist/src/resolved-case.js.map +1 -0
- package/dist/src/results-lock.js +40 -0
- package/dist/src/results-lock.js.map +1 -0
- package/dist/src/results.js +127 -0
- package/dist/src/results.js.map +1 -0
- package/dist/src/run-owner.js +283 -0
- package/dist/src/run-owner.js.map +1 -0
- package/dist/src/runner-paths.js +11 -0
- package/dist/src/runner-paths.js.map +1 -0
- package/dist/src/runtime/config.js +323 -0
- package/dist/src/runtime/config.js.map +1 -0
- package/dist/src/runtime/discovery-apple.js +65 -0
- package/dist/src/runtime/discovery-apple.js.map +1 -0
- package/dist/src/runtime/discovery-providers.js +37 -0
- package/dist/src/runtime/discovery-providers.js.map +1 -0
- package/dist/src/runtime/discovery.js +40 -0
- package/dist/src/runtime/discovery.js.map +1 -0
- package/dist/src/runtime/hook-processes.js +81 -0
- package/dist/src/runtime/hook-processes.js.map +1 -0
- package/dist/src/runtime/hooks.js +304 -0
- package/dist/src/runtime/hooks.js.map +1 -0
- package/dist/src/runtime/paths.js +244 -0
- package/dist/src/runtime/paths.js.map +1 -0
- package/dist/src/runtime/query.js +76 -0
- package/dist/src/runtime/query.js.map +1 -0
- package/dist/src/runtime/seeds.js +98 -0
- package/dist/src/runtime/seeds.js.map +1 -0
- package/dist/src/runtime/shared-cache.js +81 -0
- package/dist/src/runtime/shared-cache.js.map +1 -0
- package/dist/src/runtime/state.js +106 -0
- package/dist/src/runtime/state.js.map +1 -0
- package/dist/src/runtime-process.js +148 -0
- package/dist/src/runtime-process.js.map +1 -0
- package/dist/src/schema-types.js +2 -0
- package/dist/src/schema-types.js.map +1 -0
- package/dist/src/schema.js +16 -0
- package/dist/src/schema.js.map +1 -0
- package/dist/src/shell.js +5 -0
- package/dist/src/shell.js.map +1 -0
- package/dist/src/sqlite-lock.js +18 -0
- package/dist/src/sqlite-lock.js.map +1 -0
- package/dist/src/storage.js +21 -0
- package/dist/src/storage.js.map +1 -0
- package/dist/src/transcript-retention.js +109 -0
- package/dist/src/transcript-retention.js.map +1 -0
- package/dist/src/value-guards.js +26 -0
- package/dist/src/value-guards.js.map +1 -0
- package/docs/README.md +19 -15
- package/docs/architecture.md +141 -17
- package/docs/assets/README.md +5 -4
- package/docs/builtin-graders-v1.md +9 -2
- package/docs/codex-host.md +155 -57
- package/docs/contracts.md +2 -1
- package/docs/creating-cases.md +10 -4
- package/docs/development-cli.md +197 -24
- package/docs/documentation-quality.md +16 -5
- package/docs/evaluation-workflow.md +110 -0
- package/docs/extension-protocol-v1.md +26 -3
- package/docs/getting-started.md +29 -7
- package/docs/guide-evaluation.md +34 -2
- package/docs/identity-v1.md +29 -3
- package/docs/installing.md +35 -17
- package/docs/licensing.md +2 -2
- package/docs/native-controls.md +76 -5
- package/docs/native-hosts.md +197 -8
- package/docs/reading-results.md +102 -0
- package/docs/recipes.md +490 -0
- package/docs/releases.md +48 -13
- package/docs/report-v1.md +29 -1
- package/docs/results-v1.md +122 -4
- package/docs/run-settings-v1.md +147 -0
- package/docs/runtime-v1.md +255 -22
- package/docs/specs/repository-guide.md +1 -1
- package/docs/troubleshooting.md +12 -12
- package/docs/typescript-quality.md +106 -26
- package/examples/recipes/candidate.ts +26 -0
- package/examples/recipes/cases/coding-change.json +21 -0
- package/examples/recipes/cases/host-tools.json +21 -0
- package/examples/recipes/cases/mise.json +21 -0
- package/examples/recipes/cases/node.json +21 -0
- package/examples/recipes/cases/python-uv.json +21 -0
- package/examples/recipes/cases/semantic.json +17 -0
- package/examples/recipes/coding-change/check.ts +23 -0
- package/examples/recipes/executable.ts +19 -0
- package/examples/recipes/extension-artifacts.ts +65 -0
- package/examples/recipes/extension-methods.ts +174 -0
- package/examples/recipes/extension.ts +38 -0
- package/examples/recipes/modules.ts +9 -0
- package/examples/recipes/repository-skill/.agents/skills/startup-guide/SKILL.md +7 -0
- package/examples/recipes/repository-skill/.claude/skills/startup-guide/SKILL.md +7 -0
- package/examples/recipes/repository-skill/README.md +3 -0
- package/examples/recipes/run-inputs.ts +183 -0
- package/examples/recipes/run-options.ts +132 -0
- package/examples/recipes/run.ts +77 -0
- package/examples/recipes/runtime/empty.json +1 -0
- package/examples/recipes/runtime/goals.json +1 -0
- package/examples/recipes/runtime/mise.json +6 -0
- package/examples/recipes/runtime/path.json +1 -0
- package/examples/recipes/runtime/plugins.json +4 -0
- package/examples/recipes/runtime/python-uv.json +10 -0
- package/examples/recipes/semantic-judge.ts +29 -0
- package/examples/templates/coding/adapter.ts +17 -0
- package/examples/templates/coding/case.json +15 -0
- package/examples/templates/repository-skill/.agents/skills/startup-guide/SKILL.md +6 -0
- package/examples/templates/repository-skill/extension-data.ts +64 -0
- package/examples/templates/repository-skill/extension.ts +32 -0
- package/examples/templates/response/adapter.ts +12 -0
- package/examples/templates/response/case.json +13 -0
- package/package.json +23 -9
- package/schemas/report-v1.schema.json +9 -0
- package/schemas/run-evidence-v1.schema.json +2 -1
- package/schemas/runtime-v1.schema.json +43 -7
- package/src/advisory.ts +2 -2
- package/src/candidate-artifacts.ts +42 -11
- package/src/candidate-codex-state.ts +185 -0
- package/src/case-validation.ts +73 -50
- package/src/cli/cancellation.ts +20 -0
- package/src/cli/command.ts +87 -0
- package/src/cli/completion.ts +15 -0
- package/src/cli/demo.ts +61 -0
- package/src/cli/evaluation.ts +55 -0
- package/src/cli/extension.ts +79 -0
- package/src/cli/help.ts +199 -0
- package/src/cli/hosts.ts +55 -0
- package/src/cli/init-settings.ts +31 -0
- package/src/cli/init.ts +112 -0
- package/src/{cli-invocation.ts → cli/invocation.ts} +215 -218
- package/src/cli/loading.ts +151 -0
- package/src/cli/preparation.ts +154 -0
- package/src/cli/presentation.ts +93 -0
- package/src/cli/progress.ts +6 -0
- package/src/cli/run-manifest.ts +128 -0
- package/src/cli/run-options.ts +403 -0
- package/src/cli/run-overrides.ts +69 -0
- package/src/cli/settings.ts +154 -0
- package/src/cli/summary.ts +43 -0
- package/src/cli.ts +14 -592
- package/src/contract-types.ts +94 -0
- package/src/engine.ts +9 -5
- package/src/evaluation/advisory.ts +150 -0
- package/src/{evaluation-check-plan.ts → evaluation/check-plan.ts} +103 -67
- package/src/evaluation/collection.ts +88 -0
- package/src/evaluation/context.ts +10 -0
- package/src/{evaluation-grading-host.ts → evaluation/grading-host.ts} +5 -5
- package/src/evaluation/grading.ts +235 -0
- package/src/{evaluation-invocation-plan.ts → evaluation/invocation-plan.ts} +147 -58
- package/src/{evaluation-plan.ts → evaluation/plan.ts} +208 -141
- package/src/evaluation/progress.ts +19 -0
- package/src/evaluation/resources.ts +241 -0
- package/src/evaluation/run.ts +206 -0
- package/src/{evaluation-semantic.ts → evaluation/semantic.ts} +77 -87
- package/src/evaluation/trial.ts +800 -0
- package/src/{evaluation-types.ts → evaluation/types.ts} +79 -29
- package/src/evidence/host-results.ts +85 -0
- package/src/evidence/retained-artifacts.ts +18 -0
- package/src/evidence/trial-writer.ts +203 -0
- package/src/evidence/usage.ts +17 -0
- package/src/executable.ts +19 -0
- package/src/extension-client.ts +40 -108
- package/src/extension-session.ts +24 -48
- package/src/{advisory-fixture.ts → fixtures/advisory.ts} +12 -72
- package/src/{fixture-bin.ts → fixtures/bin.ts} +1 -1
- package/src/{fixture-cleanup.ts → fixtures/cleanup.ts} +13 -7
- package/src/fixtures/create.ts +183 -0
- package/src/{generated-fixture.ts → fixtures/generated.ts} +10 -53
- package/src/{git-hooks.ts → fixtures/git-hooks.ts} +1 -1
- package/src/{preparation.ts → fixtures/preparation.ts} +6 -16
- package/src/{repository-fixture.ts → fixtures/repository.ts} +30 -64
- package/src/{fixture-setup.ts → fixtures/setup.ts} +36 -53
- package/src/{fixture-tools.ts → fixtures/tools.ts} +1 -1
- package/src/{workspace-state.ts → fixtures/workspace-state.ts} +5 -5
- package/src/generated/cli-result.d.cts +2 -1
- package/src/generated/extension.d.cts +2 -1
- package/src/generated/report.cjs +1 -1
- package/src/generated/report.d.cts +2 -1
- package/src/generated/run-evidence.cjs +1 -1
- package/src/generated/run-evidence.d.cts +2 -1
- package/src/generated/runtime.cjs +1 -1
- package/src/generated/runtime.d.cts +2 -1
- package/src/git.ts +94 -0
- package/src/graders/git-head.ts +31 -85
- package/src/graders/shell.ts +54 -71
- package/src/host-diagnostic-redaction.ts +52 -0
- package/src/host-diagnostics.ts +331 -0
- package/src/hosts/claude-continuation.ts +57 -40
- package/src/hosts/claude-credential.ts +160 -63
- package/src/hosts/claude-events.ts +34 -18
- package/src/hosts/claude-keychain.ts +79 -21
- package/src/hosts/claude-login-recovery.ts +85 -0
- package/src/hosts/claude-login-refresh.ts +179 -0
- package/src/hosts/claude-login-store.ts +162 -0
- package/src/hosts/claude-login.ts +107 -0
- package/src/hosts/claude-native-goal.ts +15 -12
- package/src/hosts/claude-process.ts +10 -50
- package/src/hosts/claude-refresh-lock.ts +82 -0
- package/src/hosts/claude-repository-invocation.ts +9 -34
- package/src/hosts/claude-settings.ts +46 -9
- package/src/hosts/claude-skill-activations.ts +169 -0
- package/src/hosts/claude-state.ts +28 -15
- package/src/hosts/claude-tool-calls.ts +13 -10
- package/src/hosts/claude.ts +121 -81
- package/src/hosts/codex-app-server.ts +78 -52
- package/src/hosts/codex-configuration.ts +5 -10
- package/src/hosts/codex-exec-controls.ts +516 -0
- package/src/hosts/codex-marketplace.ts +12 -10
- package/src/hosts/codex-native-calls.ts +15 -122
- package/src/hosts/codex-process.ts +10 -183
- package/src/hosts/codex-profile.ts +36 -21
- package/src/hosts/codex-session-files.ts +1 -9
- package/src/hosts/codex-state.ts +134 -18
- package/src/hosts/codex.ts +210 -183
- package/src/hosts/isolation-common.ts +8 -11
- package/src/hosts/isolation-roots.ts +5 -4
- package/src/hosts/linux-sandbox.ts +15 -2
- package/src/hosts/mac-sandbox.ts +87 -18
- package/src/hosts/native-controls.ts +72 -22
- package/src/hosts/native-request.ts +55 -0
- package/src/hosts/process-lifecycle.ts +215 -49
- package/src/hosts/protected-worktrees.ts +3 -2
- package/src/hosts/skill-hook-catalog.ts +131 -0
- package/src/hosts/skill-hook-handler.ts +374 -0
- package/src/hosts/skill-hooks.ts +585 -0
- package/src/instrumentation.ts +7 -8
- package/src/native-transcript-state.ts +22 -3
- package/src/native-transcript-view.ts +52 -13
- package/src/native-transcripts.ts +26 -11
- package/src/paths.ts +22 -0
- package/src/process-lifecycle.ts +150 -0
- package/src/provenance.ts +34 -13
- package/src/prune.ts +300 -0
- package/src/report.ts +63 -36
- package/src/resolved-case.ts +1 -1
- package/src/results-lock.ts +42 -0
- package/src/results.ts +14 -18
- package/src/run-owner.ts +15 -9
- package/src/runner-paths.ts +21 -0
- package/src/{runtime-config.ts → runtime/config.ts} +160 -46
- package/src/{runtime-discovery-apple.ts → runtime/discovery-apple.ts} +3 -3
- package/src/{runtime-discovery-providers.ts → runtime/discovery-providers.ts} +3 -9
- package/src/{runtime-discovery.ts → runtime/discovery.ts} +4 -4
- package/src/{runtime-hook-processes.ts → runtime/hook-processes.ts} +3 -2
- package/src/{runtime-hooks.ts → runtime/hooks.ts} +40 -28
- package/src/{runtime-paths.ts → runtime/paths.ts} +5 -19
- package/src/{runtime-query.ts → runtime/query.ts} +5 -5
- package/src/{runtime-seeds.ts → runtime/seeds.ts} +3 -3
- package/src/runtime/shared-cache.ts +130 -0
- package/src/{runtime-state.ts → runtime/state.ts} +28 -11
- package/src/runtime-process.ts +194 -0
- package/src/schema-types.ts +73 -161
- package/src/shell.ts +4 -0
- package/src/sqlite-lock.ts +25 -0
- package/src/transcript-retention.ts +146 -0
- package/src/value-guards.ts +22 -0
- package/docs/assets/sevro-logo.png +0 -0
- package/docs/assets/sevro-symbol.svg +0 -1
- package/docs/assets/sevro-wordmark.svg +0 -1
- package/docs/cache-ownership-validation.md +0 -49
- package/docs/claude-continuation-validation.md +0 -47
- package/docs/concurrency-validation.md +0 -86
- package/docs/configuration-root-validation.md +0 -70
- package/docs/engine-slice.md +0 -114
- package/docs/evidence/runtime-startup-30/README.md +0 -185
- package/docs/evidence/runtime-startup-30/after.json +0 -319
- package/docs/evidence/runtime-startup-30/baseline-timeout.jsonl +0 -2
- package/docs/evidence/runtime-startup-30/before.json +0 -403
- package/docs/evidence/runtime-startup-30/consumer-smoke.json +0 -63
- package/docs/evidence/runtime-startup-30/full-policy-diagnostic.jsonl +0 -3
- package/docs/evidence/runtime-startup-30/minimal-policy-diagnostic.jsonl +0 -3
- package/docs/evidence/runtime-startup-30/plugin-template-assessment.json +0 -20
- package/docs/evidence/runtime-startup-30/seedless-timeout.jsonl +0 -2
- package/docs/evidence.md +0 -17
- package/docs/fixture-cleanup-validation.md +0 -73
- package/docs/guide-verification.json +0 -239
- package/src/evaluation-context.ts +0 -259
- package/src/evaluation-fixture.ts +0 -302
- package/src/evaluation-run.ts +0 -200
- package/src/evaluation-trial.ts +0 -1016
- package/src/hosts/claude-nested-skills.ts +0 -272
- package/src/hosts/codex-command-output.ts +0 -510
- package/src/hosts/codex-executor-command.ts +0 -291
- package/src/hosts/codex-skill-reads.ts +0 -783
- /package/src/{trial-scheduler.ts → evaluation/scheduler.ts} +0 -0
|
@@ -1,272 +0,0 @@
|
|
|
1
|
-
import { lstat, readFile, readdir, realpath } from "node:fs/promises";
|
|
2
|
-
import { join } from "node:path";
|
|
3
|
-
import { isRecord } from "../value-guards";
|
|
4
|
-
|
|
5
|
-
type Entry = Record<string, unknown>;
|
|
6
|
-
const MAX_BYTES = 8 * 1024 * 1024;
|
|
7
|
-
const MAX_AGENTS = 64;
|
|
8
|
-
const MAX_SKILLS = 128;
|
|
9
|
-
|
|
10
|
-
function label(value: unknown): value is string {
|
|
11
|
-
return typeof value === "string" && /^[A-Za-z0-9_.:-]{1,256}$/.test(value);
|
|
12
|
-
}
|
|
13
|
-
|
|
14
|
-
function identity(value: unknown): value is string {
|
|
15
|
-
return typeof value === "string" && /^[A-Za-z0-9_-]{1,128}$/.test(value);
|
|
16
|
-
}
|
|
17
|
-
|
|
18
|
-
function blocks(entry: Entry): Entry[] {
|
|
19
|
-
const content = isRecord(entry.message) ? entry.message.content : null;
|
|
20
|
-
return Array.isArray(content) ? content.filter(isRecord) : [];
|
|
21
|
-
}
|
|
22
|
-
|
|
23
|
-
function entries(text: string): Entry[] {
|
|
24
|
-
return text
|
|
25
|
-
.split(/\r?\n/)
|
|
26
|
-
.filter(Boolean)
|
|
27
|
-
.map((line) => {
|
|
28
|
-
const parsed: unknown = JSON.parse(line);
|
|
29
|
-
if (!isRecord(parsed)) throw new Error("invalid Claude session entry");
|
|
30
|
-
return parsed;
|
|
31
|
-
});
|
|
32
|
-
}
|
|
33
|
-
|
|
34
|
-
async function readSession(path: string): Promise<string> {
|
|
35
|
-
const file = await lstat(path);
|
|
36
|
-
if (!file.isFile() || file.isSymbolicLink() || file.size > MAX_BYTES)
|
|
37
|
-
throw new Error("Claude session file is unavailable");
|
|
38
|
-
return readFile(path, "utf8");
|
|
39
|
-
}
|
|
40
|
-
|
|
41
|
-
type NestedSkill = {
|
|
42
|
-
ancestorToolUseId: string;
|
|
43
|
-
skill: string;
|
|
44
|
-
invocation: string;
|
|
45
|
-
};
|
|
46
|
-
|
|
47
|
-
type State = {
|
|
48
|
-
sessionId: string;
|
|
49
|
-
projectDir: string;
|
|
50
|
-
seen: Set<string>;
|
|
51
|
-
calls: NestedSkill[];
|
|
52
|
-
};
|
|
53
|
-
|
|
54
|
-
async function sidecarChild(
|
|
55
|
-
state: State,
|
|
56
|
-
toolUseId: string,
|
|
57
|
-
parentAgentId: string | null,
|
|
58
|
-
): Promise<string> {
|
|
59
|
-
const directory = join(state.projectDir, state.sessionId, "subagents");
|
|
60
|
-
const names = (await readdir(directory)).filter((name) =>
|
|
61
|
-
/^agent-[A-Za-z0-9_-]+\.meta\.json$/.test(name),
|
|
62
|
-
);
|
|
63
|
-
if (names.length > MAX_AGENTS) throw new Error("excessive Claude agents");
|
|
64
|
-
const matches: string[] = [];
|
|
65
|
-
for (const name of names) {
|
|
66
|
-
const metadata: unknown = JSON.parse(
|
|
67
|
-
await readSession(join(directory, name)),
|
|
68
|
-
);
|
|
69
|
-
if (metadataMatches(metadata, toolUseId, parentAgentId))
|
|
70
|
-
matches.push(name.slice(6, -10));
|
|
71
|
-
}
|
|
72
|
-
return uniqueChildIdentity(matches);
|
|
73
|
-
}
|
|
74
|
-
|
|
75
|
-
function metadataMatches(
|
|
76
|
-
metadata: unknown,
|
|
77
|
-
toolUseId: string,
|
|
78
|
-
parentAgentId: string | null,
|
|
79
|
-
): boolean {
|
|
80
|
-
return (
|
|
81
|
-
isRecord(metadata) &&
|
|
82
|
-
metadata.toolUseId === toolUseId &&
|
|
83
|
-
(metadata.parentAgentId ?? null) === parentAgentId
|
|
84
|
-
);
|
|
85
|
-
}
|
|
86
|
-
function uniqueChildIdentity(matches: string[]): string {
|
|
87
|
-
const [id] = matches;
|
|
88
|
-
if (matches.length !== 1 || !identity(id))
|
|
89
|
-
throw new Error("Claude child identity is unavailable");
|
|
90
|
-
return id;
|
|
91
|
-
}
|
|
92
|
-
|
|
93
|
-
async function childId(
|
|
94
|
-
entry: Entry,
|
|
95
|
-
state: State,
|
|
96
|
-
toolUseId: string,
|
|
97
|
-
parentAgentId: string | null,
|
|
98
|
-
): Promise<string> {
|
|
99
|
-
const result = isRecord(entry.toolUseResult) ? entry.toolUseResult : null;
|
|
100
|
-
if (result) {
|
|
101
|
-
if (result.status !== "completed" || !identity(result.agentId))
|
|
102
|
-
throw new Error("Claude child did not complete");
|
|
103
|
-
return result.agentId;
|
|
104
|
-
}
|
|
105
|
-
return sidecarChild(state, toolUseId, parentAgentId);
|
|
106
|
-
}
|
|
107
|
-
|
|
108
|
-
async function childSession(state: State, agentId: string) {
|
|
109
|
-
const directory = join(state.projectDir, state.sessionId, "subagents");
|
|
110
|
-
const path = join(directory, `agent-${agentId}.jsonl`);
|
|
111
|
-
if ((await realpath(path)) !== path)
|
|
112
|
-
throw new Error("Claude child session is redirected");
|
|
113
|
-
return readSession(path);
|
|
114
|
-
}
|
|
115
|
-
|
|
116
|
-
function skillCall(block: Entry, ancestorToolUseId: string | null) {
|
|
117
|
-
const invocation = isRecord(block.input) ? block.input.skill : null;
|
|
118
|
-
if (!label(invocation)) throw new Error("invalid Claude Skill call");
|
|
119
|
-
return ancestorToolUseId
|
|
120
|
-
? {
|
|
121
|
-
ancestorToolUseId,
|
|
122
|
-
skill: invocation.slice(invocation.lastIndexOf(":") + 1),
|
|
123
|
-
invocation,
|
|
124
|
-
}
|
|
125
|
-
: null;
|
|
126
|
-
}
|
|
127
|
-
|
|
128
|
-
type VisitContext = {
|
|
129
|
-
state: State;
|
|
130
|
-
agentId: string | null;
|
|
131
|
-
ancestorToolUseId: string | null;
|
|
132
|
-
pending: Set<string>;
|
|
133
|
-
};
|
|
134
|
-
|
|
135
|
-
async function visit(
|
|
136
|
-
text: string,
|
|
137
|
-
state: State,
|
|
138
|
-
agentId: string | null,
|
|
139
|
-
ancestorToolUseId: string | null,
|
|
140
|
-
): Promise<void> {
|
|
141
|
-
const context: VisitContext = {
|
|
142
|
-
state,
|
|
143
|
-
agentId,
|
|
144
|
-
ancestorToolUseId,
|
|
145
|
-
pending: new Set(),
|
|
146
|
-
};
|
|
147
|
-
for (const entry of entries(text)) {
|
|
148
|
-
if (entry.type !== "assistant" && entry.type !== "user") continue;
|
|
149
|
-
await visitEntry(entry, context);
|
|
150
|
-
}
|
|
151
|
-
if (context.pending.size) throw new Error("incomplete Claude Agent graph");
|
|
152
|
-
}
|
|
153
|
-
|
|
154
|
-
async function visitEntry(entry: Entry, context: VisitContext): Promise<void> {
|
|
155
|
-
if (
|
|
156
|
-
entry.sessionId !== context.state.sessionId ||
|
|
157
|
-
(entry.agentId ?? null) !== context.agentId
|
|
158
|
-
)
|
|
159
|
-
throw new Error("Claude session identity mismatch");
|
|
160
|
-
for (const block of blocks(entry)) await visitBlock(entry, block, context);
|
|
161
|
-
}
|
|
162
|
-
|
|
163
|
-
async function visitBlock(
|
|
164
|
-
entry: Entry,
|
|
165
|
-
block: Entry,
|
|
166
|
-
context: VisitContext,
|
|
167
|
-
): Promise<void> {
|
|
168
|
-
if (entry.type === "assistant" && block.type === "tool_use")
|
|
169
|
-
assistantBlock(block, context);
|
|
170
|
-
if (acceptedChildResult(block, entry, context.pending))
|
|
171
|
-
await visitChild(entry, block, context);
|
|
172
|
-
}
|
|
173
|
-
|
|
174
|
-
function acceptedChildResult(
|
|
175
|
-
block: Entry,
|
|
176
|
-
entry: Entry,
|
|
177
|
-
pending: Set<string>,
|
|
178
|
-
): block is Entry & { tool_use_id: string } {
|
|
179
|
-
return (
|
|
180
|
-
entry.type === "user" &&
|
|
181
|
-
block.type === "tool_result" &&
|
|
182
|
-
identity(block.tool_use_id) &&
|
|
183
|
-
pending.has(block.tool_use_id)
|
|
184
|
-
);
|
|
185
|
-
}
|
|
186
|
-
|
|
187
|
-
function assistantBlock(block: Entry, context: VisitContext): void {
|
|
188
|
-
if (block.name === "Skill") retainSkill(block, context);
|
|
189
|
-
if (block.name === "Agent" || block.name === "Task")
|
|
190
|
-
registerAgent(block, context.pending);
|
|
191
|
-
}
|
|
192
|
-
|
|
193
|
-
function retainSkill(block: Entry, context: VisitContext): void {
|
|
194
|
-
const call = skillCall(block, context.ancestorToolUseId);
|
|
195
|
-
if (call) context.state.calls.push(call);
|
|
196
|
-
if (context.state.calls.length > MAX_SKILLS)
|
|
197
|
-
throw new Error("excessive Claude Skill calls");
|
|
198
|
-
}
|
|
199
|
-
|
|
200
|
-
function registerAgent(block: Entry, pending: Set<string>): void {
|
|
201
|
-
if (!identity(block.id) || pending.has(block.id))
|
|
202
|
-
throw new Error("invalid Claude Agent call");
|
|
203
|
-
pending.add(block.id);
|
|
204
|
-
}
|
|
205
|
-
|
|
206
|
-
async function visitChild(
|
|
207
|
-
entry: Entry,
|
|
208
|
-
block: Entry & { tool_use_id: string },
|
|
209
|
-
context: VisitContext,
|
|
210
|
-
): Promise<void> {
|
|
211
|
-
if (block.is_error === true) throw new Error("Claude child call failed");
|
|
212
|
-
const { state, agentId, ancestorToolUseId, pending } = context;
|
|
213
|
-
const child = await childId(entry, state, block.tool_use_id, agentId);
|
|
214
|
-
if (state.seen.has(child) || state.seen.size >= MAX_AGENTS)
|
|
215
|
-
throw new Error("repeated Claude child");
|
|
216
|
-
state.seen.add(child);
|
|
217
|
-
await visit(
|
|
218
|
-
await childSession(state, child),
|
|
219
|
-
state,
|
|
220
|
-
child,
|
|
221
|
-
ancestorToolUseId ?? block.tool_use_id,
|
|
222
|
-
);
|
|
223
|
-
pending.delete(block.tool_use_id);
|
|
224
|
-
}
|
|
225
|
-
|
|
226
|
-
export async function claudeNestedSkillsObservation(
|
|
227
|
-
stream: string,
|
|
228
|
-
configRoot: string,
|
|
229
|
-
workspace: string,
|
|
230
|
-
) {
|
|
231
|
-
const unavailable = {
|
|
232
|
-
id: "sevro.claude.nested-skills" as const,
|
|
233
|
-
completeness: "partial" as const,
|
|
234
|
-
data: { method: "native_session_graph", calls: [] as NestedSkill[] },
|
|
235
|
-
};
|
|
236
|
-
try {
|
|
237
|
-
const sessions = [
|
|
238
|
-
...new Set(
|
|
239
|
-
entries(stream)
|
|
240
|
-
.filter((entry) => entry.type === "result")
|
|
241
|
-
.map((entry) => entry.session_id),
|
|
242
|
-
),
|
|
243
|
-
];
|
|
244
|
-
if (sessions.length !== 1 || !identity(sessions[0])) return unavailable;
|
|
245
|
-
const canonicalWorkspace = await realpath(workspace);
|
|
246
|
-
const canonicalConfigRoot = await realpath(configRoot);
|
|
247
|
-
const projectDir = join(
|
|
248
|
-
canonicalConfigRoot,
|
|
249
|
-
"projects",
|
|
250
|
-
canonicalWorkspace.replace(/[^A-Za-z0-9]/g, "-"),
|
|
251
|
-
);
|
|
252
|
-
const state: State = {
|
|
253
|
-
sessionId: sessions[0],
|
|
254
|
-
projectDir,
|
|
255
|
-
seen: new Set(),
|
|
256
|
-
calls: [],
|
|
257
|
-
};
|
|
258
|
-
await visit(
|
|
259
|
-
await readSession(join(projectDir, `${state.sessionId}.jsonl`)),
|
|
260
|
-
state,
|
|
261
|
-
null,
|
|
262
|
-
null,
|
|
263
|
-
);
|
|
264
|
-
return {
|
|
265
|
-
id: "sevro.claude.nested-skills" as const,
|
|
266
|
-
completeness: "complete" as const,
|
|
267
|
-
data: { method: "native_session_graph", calls: state.calls },
|
|
268
|
-
};
|
|
269
|
-
} catch {
|
|
270
|
-
return unavailable;
|
|
271
|
-
}
|
|
272
|
-
}
|