parallax-evals 2.0.0.dev2__tar.gz → 2.0.0.dev4__tar.gz
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.
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/PKG-INFO +2 -2
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/README.md +1 -1
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/docs/methodology.md +2 -2
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/docs/roadmap.md +1 -1
- parallax_evals-2.0.0.dev4/parallax/__init__.py +1 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/base.Dockerfile +4 -5
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/box.py +45 -26
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/extensions.py +3 -8
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/harnesses/base.py +2 -6
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/harnesses/oracle.py +2 -6
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/runner.py +3 -7
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/semantic.py +3 -2
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/track/sink.py +2 -2
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/service.py +3 -7
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-two-turn-carry/solution/oracle.yaml +1 -1
- parallax_evals-2.0.0.dev4/tasks/conformance/conf-two-turn-carry/steps/turn-01/instruction.md +2 -0
- parallax_evals-2.0.0.dev4/tasks/conformance/conf-two-turn-carry/steps/turn-02/instruction.md +1 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-two-turn-carry/task.toml +2 -2
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-two-turn-carry/tests/checks.yaml +1 -1
- parallax_evals-2.0.0.dev4/tasks/conformance/conf-two-turn-carry/tests/mutations.yaml +9 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_semantic_grading.py +3 -1
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_track.py +3 -0
- parallax_evals-2.0.0.dev2/parallax/__init__.py +0 -1
- parallax_evals-2.0.0.dev2/tasks/conformance/conf-two-turn-carry/steps/turn-01/instruction.md +0 -2
- parallax_evals-2.0.0.dev2/tasks/conformance/conf-two-turn-carry/steps/turn-02/instruction.md +0 -1
- parallax_evals-2.0.0.dev2/tasks/conformance/conf-two-turn-carry/tests/mutations.yaml +0 -9
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.agents/skills/task-authoring/SKILL.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.dockerignore +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.github/workflows/ci.yml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.github/workflows/release.yml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.gitignore +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.pre-commit-config.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/LICENSE +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/agents.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/gate.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/lenses/arbitrary-caps.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/lenses/eval-fairness.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/lenses/unmeasured-figures.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/script-tests/no-long-dashes/clean/expected +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/script-tests/no-long-dashes/clean/head/docs/guide.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/script-tests/no-long-dashes/clean/head/tasks/t/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/script-tests/no-long-dashes/flags/expected +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/script-tests/no-long-dashes/flags/head/docs/guide.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/script-tests/one-langfuse-client/clean/expected +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/script-tests/one-langfuse-client/clean/head/parallax/track/sink.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/script-tests/one-langfuse-client/flags/expected +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/script-tests/one-langfuse-client/flags/head/parallax/track/upload.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/script-tests/pinned-actions/clean/expected +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/script-tests/pinned-actions/clean/head/.github/workflows/ci.yml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/script-tests/pinned-actions/flags/expected +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/script-tests/pinned-actions/flags/head/parallax/ci.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/scripts/no-long-dashes +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/scripts/one-langfuse-client +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/scripts/pinned-actions +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/CODE_OF_CONDUCT.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/CONTRIBUTING.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/LICENSE +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/LICENSE-DATA +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/NOTICE +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/SECURITY.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/THIRD_PARTY_DATA.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/banner.svg +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/OFL-Roboto.txt +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/README.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/contact-sheet.html +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/favicon.ico +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/generate.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/icon-128.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/icon-16.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/icon-180.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/icon-192.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/icon-256.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/icon-32.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/icon-512.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/icon-64.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/logo-lockup-800.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/logo-lockup-dark-800.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/logo-lockup-dark.svg +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/logo-lockup.svg +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/logo.svg +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/mark-256.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/mark-512.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/mark-dark-512.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/mark-dark.svg +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/mark.svg +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/wordmark-800.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/wordmark-dark-800.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/wordmark-dark.svg +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/wordmark.svg +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/calibration/README.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/docs/adapters.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/docs/authoring.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/docs/jev-judge.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/docs/quickstart.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/docs/taxonomy.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/docs/tracking.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/docs/troubleshooting.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/box.Dockerfile +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/ci.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/cli.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/controls.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/desktop.Dockerfile +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/desktop.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/harnesses/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/harnesses/manifest.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/harnesses/manifests/claude-code.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/harnesses/manifests/codex.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/harnesses/manifests/grok-build.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/harnesses/manifests/hermes.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/harnesses/manifests/openclaw.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/harnesses/null.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/harnesses/parsers.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/images.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/jev.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/judge.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/judging.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/match.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/meter.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/models.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/plugins.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/protocols.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/provenance.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/px_clock.c +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/stats.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/task.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/track/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/track/compare.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/track/projection.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/track/rows.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/track/upload.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/verify.Dockerfile +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/__main__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/apps.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/client.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/core.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/server.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/analytics.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/calendar.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/chat.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/common.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/contacts.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/drive.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/helpdesk.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/mail.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/payments.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/recurrence.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/scheduler.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/sheets.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/tables.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/tracker.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/user.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/web.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/parallax/world/services/wiki.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/pyproject.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/review/drafts/README.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/scripts/scan_public.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-fault-then-retry/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-fault-then-retry/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-fault-then-retry/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-fault-then-retry/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-fault-then-retry/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-fault-then-retry/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-mcp-echo/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-mcp-echo/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-mcp-echo/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-mcp-echo/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-mcp-echo/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-mcp-echo/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-memory-recall/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-memory-recall/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-memory-recall/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-memory-recall/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-memory-recall/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-memory-recall/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-memory-recall/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-two-turn-carry/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-workspace-edit-verify/environment/files/app.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-workspace-edit-verify/environment/files/test_app.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-workspace-edit-verify/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-workspace-edit-verify/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-workspace-edit-verify/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-workspace-edit-verify/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-workspace-edit-verify/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/conf-workspace-edit-verify/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/conformance/suite.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-add-helper-and-wire/environment/files/jotter/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-add-helper-and-wire/environment/files/jotter/__main__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-add-helper-and-wire/environment/files/jotter/cli.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-add-helper-and-wire/environment/files/jotter/text.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-add-helper-and-wire/environment/files/tests/test_cli.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-add-helper-and-wire/environment/files/tests/test_text.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-add-helper-and-wire/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-add-helper-and-wire/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-add-helper-and-wire/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-add-helper-and-wire/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-add-helper-and-wire/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-add-helper-and-wire/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-dependency-api-change/environment/files/requirements.txt +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-dependency-api-change/environment/files/shopfront/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-dependency-api-change/environment/files/shopfront/export.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-dependency-api-change/environment/files/shopfront/receipt.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-dependency-api-change/environment/files/tests/test_shopfront.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-dependency-api-change/environment/files/vendor/tinyfmt-1.4.0/tinyfmt/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-dependency-api-change/environment/files/vendor/tinyfmt-2.0.0/CHANGELOG.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-dependency-api-change/environment/files/vendor/tinyfmt-2.0.0/tinyfmt/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-dependency-api-change/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-dependency-api-change/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-dependency-api-change/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-dependency-api-change/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-dependency-api-change/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-dependency-api-change/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-extract-module/environment/files/ingest/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-extract-module/environment/files/ingest/loader.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-extract-module/environment/files/ingest/summary.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-extract-module/environment/files/ingest/utils.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-extract-module/environment/files/tests/test_ingest.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-extract-module/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-extract-module/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-extract-module/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-extract-module/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-extract-module/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-extract-module/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-fix-failing-test/environment/files/pricing/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-fix-failing-test/environment/files/pricing/basket.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-fix-failing-test/environment/files/pricing/ops.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-fix-failing-test/environment/files/tests/test_basket.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-fix-failing-test/environment/files/tests/test_ops.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-fix-failing-test/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-fix-failing-test/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-fix-failing-test/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-fix-failing-test/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-fix-failing-test/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-fix-failing-test/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-followup-keeps-constraint/environment/files/feedsync/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-followup-keeps-constraint/environment/files/feedsync/dates.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-followup-keeps-constraint/environment/files/feedsync/rows.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-followup-keeps-constraint/environment/files/pyproject.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-followup-keeps-constraint/environment/files/tests/test_dates.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-followup-keeps-constraint/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-followup-keeps-constraint/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-followup-keeps-constraint/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-followup-keeps-constraint/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-followup-keeps-constraint/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-followup-keeps-constraint/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-followup-keeps-constraint/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-locate-config/environment/files/dispatch/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-locate-config/environment/files/dispatch/config/defaults.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-locate-config/environment/files/dispatch/settings.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-locate-config/environment/files/dispatch/worker.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-locate-config/environment/files/tests/test_worker.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-locate-config/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-locate-config/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-locate-config/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-locate-config/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-locate-config/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-locate-config/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-missing-symbol-reported/environment/files/payouts/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-missing-symbol-reported/environment/files/payouts/form.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-missing-symbol-reported/environment/files/payouts/validation.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-missing-symbol-reported/environment/files/tests/test_form.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-missing-symbol-reported/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-missing-symbol-reported/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-missing-symbol-reported/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-missing-symbol-reported/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-missing-symbol-reported/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-missing-symbol-reported/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-multifile-rename/environment/files/ledger/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-multifile-rename/environment/files/ledger/formatting.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-multifile-rename/environment/files/ledger/report.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-multifile-rename/environment/files/ledger/source.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-multifile-rename/environment/files/ledger/sync.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-multifile-rename/environment/files/tests/conftest.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-multifile-rename/environment/files/tests/test_ledger.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-multifile-rename/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-multifile-rename/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-multifile-rename/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-multifile-rename/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-multifile-rename/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-multifile-rename/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-one-line-change/environment/files/relay/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-one-line-change/environment/files/relay/app.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-one-line-change/environment/files/relay/settings.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-one-line-change/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-one-line-change/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-one-line-change/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-one-line-change/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-one-line-change/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-one-line-change/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-pasted-line-numbers/environment/files/budget/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-pasted-line-numbers/environment/files/budget/limits.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-pasted-line-numbers/environment/files/tests/test_limits.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-pasted-line-numbers/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-pasted-line-numbers/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-pasted-line-numbers/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-pasted-line-numbers/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-pasted-line-numbers/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-pasted-line-numbers/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-project-relative-question/environment/files/src/worker.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-project-relative-question/environment/files/svc/README.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-project-relative-question/environment/files/svc/src/worker.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-project-relative-question/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-project-relative-question/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-project-relative-question/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-project-relative-question/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-project-relative-question/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-project-relative-question/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-project-relative-question/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-review-no-changes/environment/files/invoicing/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-review-no-changes/environment/files/invoicing/late_fees.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-review-no-changes/environment/files/invoicing/statement.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-review-no-changes/environment/files/tests/test_statement.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-review-no-changes/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-review-no-changes/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-review-no-changes/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-review-no-changes/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-review-no-changes/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-review-no-changes/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-scoped-timeout-edit/environment/files/clients/__init__.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-scoped-timeout-edit/environment/files/clients/base.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-scoped-timeout-edit/environment/files/clients/billing.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-scoped-timeout-edit/environment/files/clients/inventory.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-scoped-timeout-edit/environment/files/clients/shipping.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-scoped-timeout-edit/environment/files/tests/test_clients.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-scoped-timeout-edit/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-scoped-timeout-edit/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-scoped-timeout-edit/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-scoped-timeout-edit/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-scoped-timeout-edit/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-scoped-timeout-edit/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-aws-sso-device-code/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-aws-sso-device-code/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-aws-sso-device-code/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-aws-sso-device-code/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-aws-sso-device-code/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-aws-sso-device-code/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-az-login/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-az-login/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-az-login/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-az-login/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-az-login/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-az-login/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-gcloud-login/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-gcloud-login/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-gcloud-login/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-gcloud-login/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-gcloud-login/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-gcloud-login/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-npm-login/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-npm-login/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-npm-login/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-npm-login/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-npm-login/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/code-terminal-npm-login/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-attached-invoice-to-sheet/environment/files/attachments/invoices.csv +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-attached-invoice-to-sheet/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-attached-invoice-to-sheet/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-attached-invoice-to-sheet/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-attached-invoice-to-sheet/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-attached-invoice-to-sheet/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-attached-invoice-to-sheet/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-brief-from-packet/environment/files/research-packet.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-brief-from-packet/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-brief-from-packet/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-brief-from-packet/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-brief-from-packet/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-brief-from-packet/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-brief-from-packet/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-csv-clean-then-counts/environment/files/leads.csv +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-csv-clean-then-counts/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-csv-clean-then-counts/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-csv-clean-then-counts/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-csv-clean-then-counts/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-csv-clean-then-counts/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-csv-clean-then-counts/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-desktop-fill-expense-form/environment/files/app.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-desktop-fill-expense-form/environment/files/history.json +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-desktop-fill-expense-form/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-desktop-fill-expense-form/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-desktop-fill-expense-form/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-desktop-fill-expense-form/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-desktop-fill-expense-form/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-desktop-fill-expense-form/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-house-rules-obeyed/environment/files/AGENTS.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-house-rules-obeyed/environment/files/notes/this-week.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-house-rules-obeyed/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-house-rules-obeyed/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-house-rules-obeyed/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-house-rules-obeyed/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-house-rules-obeyed/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-house-rules-obeyed/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Desktop/todo.txt +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Downloads/IMG_2041.JPG +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Downloads/bank-export.csv +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Downloads/budget-2026.xlsx +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Downloads/cv-2026.docx +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Downloads/holiday-map.png +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Downloads/invoice-0612.pdf +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Downloads/meeting-notes.txt +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Downloads/photos-backup.zip +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Downloads/podcast-ep12.mp3 +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Downloads/project-files.zip +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Downloads/receipt-scan.jpg +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Downloads/screen-recording.mp4 +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Downloads/tenancy-agreement.pdf +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/environment/files/Downloads/voice-memo.m4a +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-organise-downloads/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-repeat-is-idempotent/environment/files/attachments/sales-2026-06.csv +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-repeat-is-idempotent/environment/files/data/notes.txt +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-repeat-is-idempotent/environment/files/data/sales-2026-03.csv +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-repeat-is-idempotent/environment/files/data/sales-2026-04.csv +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-repeat-is-idempotent/environment/files/data/sales-2026-05.csv +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-repeat-is-idempotent/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-repeat-is-idempotent/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-repeat-is-idempotent/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-repeat-is-idempotent/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-repeat-is-idempotent/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-repeat-is-idempotent/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-repeat-is-idempotent/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-revise-deck-html/environment/files/deck.html +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-revise-deck-html/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-revise-deck-html/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-revise-deck-html/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-revise-deck-html/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-revise-deck-html/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-revise-deck-html/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-transport-summary/environment/files/journeys-by-type.csv +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-transport-summary/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-transport-summary/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-transport-summary/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-transport-summary/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-transport-summary/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/files-transport-summary/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-agent-operating-contract/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-agent-operating-contract/steps/configure/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-agent-operating-contract/steps/decision/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-agent-operating-contract/steps/detail/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-agent-operating-contract/steps/status/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-agent-operating-contract/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-agent-operating-contract/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-agent-operating-contract/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-agent-operating-contract/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-chat-no-record-says-so/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-chat-no-record-says-so/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-chat-no-record-says-so/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-chat-no-record-says-so/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-chat-no-record-says-so/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-chat-no-record-says-so/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-constraint-carried/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-constraint-carried/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-constraint-carried/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-constraint-carried/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-constraint-carried/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-constraint-carried/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-constraint-carried/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-correction-supersedes/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-correction-supersedes/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-correction-supersedes/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-correction-supersedes/steps/turn-03/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-correction-supersedes/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-correction-supersedes/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-correction-supersedes/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-correction-supersedes/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-forget-on-request/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-forget-on-request/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-forget-on-request/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-forget-on-request/steps/turn-03/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-forget-on-request/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-forget-on-request/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-forget-on-request/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-forget-on-request/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-preference-applied/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-preference-applied/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-preference-applied/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-preference-applied/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-preference-applied/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-preference-applied/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-preference-applied/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-recall-agreed-figure/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-recall-agreed-figure/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-recall-agreed-figure/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-recall-agreed-figure/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-recall-agreed-figure/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-recall-agreed-figure/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/memory-recall-agreed-figure/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-calendar-cancel-one-occurrence/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-calendar-cancel-one-occurrence/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-calendar-cancel-one-occurrence/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-calendar-cancel-one-occurrence/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-calendar-cancel-one-occurrence/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-calendar-cancel-one-occurrence/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-chat-private-number/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-chat-private-number/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-chat-private-number/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-chat-private-number/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-chat-private-number/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-chat-private-number/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-contacts-right-namesake/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-contacts-right-namesake/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-contacts-right-namesake/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-contacts-right-namesake/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-contacts-right-namesake/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-contacts-right-namesake/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-desktop-settings-toggle-and-choose/environment/files/app.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-desktop-settings-toggle-and-choose/environment/files/preferences.json +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-desktop-settings-toggle-and-choose/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-desktop-settings-toggle-and-choose/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-desktop-settings-toggle-and-choose/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-desktop-settings-toggle-and-choose/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-desktop-settings-toggle-and-choose/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-desktop-settings-toggle-and-choose/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-drive-share-single-file/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-drive-share-single-file/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-drive-share-single-file/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-drive-share-single-file/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-drive-share-single-file/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-drive-share-single-file/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-mail-archive-newsletters/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-mail-archive-newsletters/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-mail-archive-newsletters/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-mail-archive-newsletters/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-mail-archive-newsletters/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-mail-archive-newsletters/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-sheets-fix-one-cell/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-sheets-fix-one-cell/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-sheets-fix-one-cell/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-sheets-fix-one-cell/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-sheets-fix-one-cell/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-sheets-fix-one-cell/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-tables-qualified-only/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-tables-qualified-only/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-tables-qualified-only/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-tables-qualified-only/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-tables-qualified-only/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-tables-qualified-only/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-tracker-close-mine-only/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-tracker-close-mine-only/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-tracker-close-mine-only/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-tracker-close-mine-only/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-tracker-close-mine-only/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/precision-tracker-close-mine-only/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-auth-expired-honest/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-auth-expired-honest/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-auth-expired-honest/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-auth-expired-honest/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-auth-expired-honest/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-auth-expired-honest/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-bulk-move-partial-failure/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-bulk-move-partial-failure/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-bulk-move-partial-failure/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-bulk-move-partial-failure/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-bulk-move-partial-failure/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-bulk-move-partial-failure/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-chat-send-fails-then-retry/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-chat-send-fails-then-retry/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-chat-send-fails-then-retry/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-chat-send-fails-then-retry/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-chat-send-fails-then-retry/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-chat-send-fails-then-retry/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-corrupt-success-readback/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-corrupt-success-readback/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-corrupt-success-readback/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-corrupt-success-readback/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-corrupt-success-readback/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-corrupt-success-readback/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-missing-file-reported/environment/files/notes/q2-plan.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-missing-file-reported/environment/files/notes/q3-roadmap-draft.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-missing-file-reported/environment/files/notes/standup.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-missing-file-reported/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-missing-file-reported/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-missing-file-reported/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-missing-file-reported/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-missing-file-reported/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-missing-file-reported/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-rate-limited-bulk/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-rate-limited-bulk/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-rate-limited-bulk/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-rate-limited-bulk/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-rate-limited-bulk/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-rate-limited-bulk/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-stale-id/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-stale-id/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-stale-id/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-stale-id/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-stale-id/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-stale-id/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-stale-id/world/patches/doc-migrated.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-stale-id/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-timeout-no-double-charge/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-timeout-no-double-charge/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-timeout-no-double-charge/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-timeout-no-double-charge/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-timeout-no-double-charge/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-timeout-no-double-charge/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-validation-then-fix/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-validation-then-fix/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-validation-then-fix/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-validation-then-fix/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-validation-then-fix/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/recovery-validation-then-fix/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-capability-absent-says-so/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-capability-absent-says-so/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-capability-absent-says-so/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-capability-absent-says-so/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-capability-absent-says-so/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-capability-absent-says-so/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-chat-thread-summary-faithful/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-chat-thread-summary-faithful/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-chat-thread-summary-faithful/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-chat-thread-summary-faithful/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-chat-thread-summary-faithful/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-chat-thread-summary-faithful/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-cites-real-urls/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-cites-real-urls/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-cites-real-urls/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-cites-real-urls/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-cites-real-urls/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-cites-real-urls/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-compare-four-vendors/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-compare-four-vendors/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-compare-four-vendors/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-compare-four-vendors/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-compare-four-vendors/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-compare-four-vendors/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-discloses-missing-evidence/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-discloses-missing-evidence/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-discloses-missing-evidence/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-discloses-missing-evidence/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-discloses-missing-evidence/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-discloses-missing-evidence/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-explicit-pricing-terms/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-explicit-pricing-terms/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-explicit-pricing-terms/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-explicit-pricing-terms/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-explicit-pricing-terms/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-explicit-pricing-terms/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-grounded-chart/environment/files/notes/q2-review.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-grounded-chart/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-grounded-chart/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-grounded-chart/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-grounded-chart/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-grounded-chart/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-grounded-chart/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-grounded-images/environment/files/newsletter-style.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-grounded-images/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-grounded-images/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-grounded-images/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-grounded-images/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-grounded-images/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-grounded-images/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-mixed-pricing-terms/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-mixed-pricing-terms/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-mixed-pricing-terms/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-mixed-pricing-terms/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-mixed-pricing-terms/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-mixed-pricing-terms/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-newest-source-wins/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-newest-source-wins/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-newest-source-wins/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-newest-source-wins/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-newest-source-wins/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-newest-source-wins/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-single-synthesis/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-single-synthesis/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-single-synthesis/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-single-synthesis/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-single-synthesis/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-single-synthesis/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-structured-api-over-pages/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-structured-api-over-pages/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-structured-api-over-pages/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-structured-api-over-pages/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-structured-api-over-pages/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/research-structured-api-over-pages/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-already-done-noop/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-already-done-noop/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-already-done-noop/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-already-done-noop/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-already-done-noop/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-already-done-noop/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-ambiguous-brief-acts/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-ambiguous-brief-acts/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-ambiguous-brief-acts/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-ambiguous-brief-acts/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-ambiguous-brief-acts/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-ambiguous-brief-acts/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-clarify-missing-details/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-clarify-missing-details/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-clarify-missing-details/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-clarify-missing-details/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-clarify-missing-details/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-clarify-missing-details/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-no-reask-given-details/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-no-reask-given-details/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-no-reask-given-details/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-no-reask-given-details/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-no-reask-given-details/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-no-reask-given-details/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-passing-mention-no-action/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-passing-mention-no-action/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-passing-mention-no-action/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-passing-mention-no-action/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-passing-mention-no-action/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-chat-passing-mention-no-action/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-clean-up-crm-vague/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-clean-up-crm-vague/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-clean-up-crm-vague/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-clean-up-crm-vague/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-clean-up-crm-vague/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-clean-up-crm-vague/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-desktop-destructive-confirm/environment/files/app.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-desktop-destructive-confirm/environment/files/collection.json +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-desktop-destructive-confirm/environment/files/deletions.json +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-desktop-destructive-confirm/environment/files/records.json +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-desktop-destructive-confirm/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-desktop-destructive-confirm/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-desktop-destructive-confirm/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-desktop-destructive-confirm/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-desktop-destructive-confirm/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-desktop-destructive-confirm/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-draft-not-send/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-draft-not-send/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-draft-not-send/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-draft-not-send/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-draft-not-send/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-draft-not-send/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-naming-choice-asks/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-naming-choice-asks/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-naming-choice-asks/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-naming-choice-asks/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-naming-choice-asks/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-naming-choice-asks/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-naming-choice-asks/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-proposes-watch-first/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-proposes-watch-first/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-proposes-watch-first/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-proposes-watch-first/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-proposes-watch-first/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-proposes-watch-first/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-proposes-watch-first/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-quick-answer-no-side-effects/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-quick-answer-no-side-effects/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-quick-answer-no-side-effects/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-quick-answer-no-side-effects/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-quick-answer-no-side-effects/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-quick-answer-no-side-effects/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-read-only-question/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-read-only-question/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-read-only-question/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-read-only-question/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-read-only-question/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-read-only-question/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-terminal-plain-commands/environment/files/logs/app.log +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-terminal-plain-commands/environment/files/logs/archive/app-2026-05.log +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-terminal-plain-commands/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-terminal-plain-commands/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-terminal-plain-commands/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-terminal-plain-commands/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-terminal-plain-commands/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-terminal-plain-commands/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-unavailable-service-no-substitute/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-unavailable-service-no-substitute/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-unavailable-service-no-substitute/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-unavailable-service-no-substitute/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-unavailable-service-no-substitute/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/restraint-unavailable-service-no-substitute/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-analytics-activation-drop/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-analytics-activation-drop/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-analytics-activation-drop/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-analytics-activation-drop/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-analytics-activation-drop/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-analytics-activation-drop/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-bug-reports-dedupe-to-tracker/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-bug-reports-dedupe-to-tracker/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-bug-reports-dedupe-to-tracker/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-bug-reports-dedupe-to-tracker/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-bug-reports-dedupe-to-tracker/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-bug-reports-dedupe-to-tracker/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-calendar-move-then-correct/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-calendar-move-then-correct/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-calendar-move-then-correct/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-calendar-move-then-correct/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-calendar-move-then-correct/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-calendar-move-then-correct/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-calendar-move-then-correct/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-calendar-mutual-slot/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-calendar-mutual-slot/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-calendar-mutual-slot/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-calendar-mutual-slot/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-calendar-mutual-slot/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-calendar-mutual-slot/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-catch-up-across-apps/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-catch-up-across-apps/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-catch-up-across-apps/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-catch-up-across-apps/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-catch-up-across-apps/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-catch-up-across-apps/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-channel-create-with-work/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-channel-create-with-work/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-channel-create-with-work/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-channel-create-with-work/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-channel-create-with-work/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-channel-create-with-work/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-chat-reply-in-thread/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-chat-reply-in-thread/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-chat-reply-in-thread/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-chat-reply-in-thread/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-chat-reply-in-thread/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-chat-reply-in-thread/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-churn-risk-across-billing-support/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-churn-risk-across-billing-support/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-churn-risk-across-billing-support/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-churn-risk-across-billing-support/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-churn-risk-across-billing-support/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-churn-risk-across-billing-support/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-commission-with-clawback/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-commission-with-clawback/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-commission-with-clawback/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-commission-with-clawback/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-commission-with-clawback/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-commission-with-clawback/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-commission-with-clawback/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-crm-at-risk-deals/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-crm-at-risk-deals/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-crm-at-risk-deals/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-crm-at-risk-deals/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-crm-at-risk-deals/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-crm-at-risk-deals/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-crm-from-meeting-notes/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-crm-from-meeting-notes/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-crm-from-meeting-notes/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-crm-from-meeting-notes/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-crm-from-meeting-notes/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-crm-from-meeting-notes/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-customer-call-prep-namesake/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-customer-call-prep-namesake/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-customer-call-prep-namesake/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-customer-call-prep-namesake/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-customer-call-prep-namesake/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-customer-call-prep-namesake/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-drive-share-then-restrict/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-drive-share-then-restrict/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-drive-share-then-restrict/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-drive-share-then-restrict/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-drive-share-then-restrict/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-drive-share-then-restrict/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-drive-share-then-restrict/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-forecast-from-wiki-method/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-forecast-from-wiki-method/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-forecast-from-wiki-method/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-forecast-from-wiki-method/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-forecast-from-wiki-method/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-forecast-from-wiki-method/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-helpdesk-answer-from-wiki/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-helpdesk-answer-from-wiki/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-helpdesk-answer-from-wiki/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-helpdesk-answer-from-wiki/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-helpdesk-answer-from-wiki/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-helpdesk-answer-from-wiki/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-helpdesk-escalate-reply-close/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-helpdesk-escalate-reply-close/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-helpdesk-escalate-reply-close/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-helpdesk-escalate-reply-close/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-helpdesk-escalate-reply-close/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-helpdesk-escalate-reply-close/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-incident-postmortem-draft/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-incident-postmortem-draft/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-incident-postmortem-draft/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-incident-postmortem-draft/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-incident-postmortem-draft/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-incident-postmortem-draft/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-investor-update-like-last-month/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-investor-update-like-last-month/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-investor-update-like-last-month/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-investor-update-like-last-month/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-investor-update-like-last-month/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-investor-update-like-last-month/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-invoice-chase-reconcile/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-invoice-chase-reconcile/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-invoice-chase-reconcile/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-invoice-chase-reconcile/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-invoice-chase-reconcile/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-invoice-chase-reconcile/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-kpi-sheet-source-of-truth/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-kpi-sheet-source-of-truth/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-kpi-sheet-source-of-truth/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-kpi-sheet-source-of-truth/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-kpi-sheet-source-of-truth/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-kpi-sheet-source-of-truth/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-launch-go-no-go/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-launch-go-no-go/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-launch-go-no-go/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-launch-go-no-go/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-launch-go-no-go/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-launch-go-no-go/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-mail-amend-then-send/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-mail-amend-then-send/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-mail-amend-then-send/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-mail-amend-then-send/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-mail-amend-then-send/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-mail-amend-then-send/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-mail-amend-then-send/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-mail-vip-scope-correction/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-mail-vip-scope-correction/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-mail-vip-scope-correction/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-mail-vip-scope-correction/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-mail-vip-scope-correction/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-mail-vip-scope-correction/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-mail-vip-scope-correction/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-offsite-rsvp-headcount/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-offsite-rsvp-headcount/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-offsite-rsvp-headcount/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-offsite-rsvp-headcount/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-offsite-rsvp-headcount/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-offsite-rsvp-headcount/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-onboard-new-hire-by-role/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-onboard-new-hire-by-role/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-onboard-new-hire-by-role/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-onboard-new-hire-by-role/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-onboard-new-hire-by-role/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-onboard-new-hire-by-role/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-receipts-to-sheet/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-receipts-to-sheet/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-receipts-to-sheet/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-receipts-to-sheet/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-receipts-to-sheet/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-receipts-to-sheet/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-retention-by-plan-excludes-internal/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-retention-by-plan-excludes-internal/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-retention-by-plan-excludes-internal/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-retention-by-plan-excludes-internal/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-retention-by-plan-excludes-internal/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-retention-by-plan-excludes-internal/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sheets-append-then-total/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sheets-append-then-total/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sheets-append-then-total/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sheets-append-then-total/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sheets-append-then-total/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sheets-append-then-total/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sheets-append-then-total/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sheets-budget-variance/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sheets-budget-variance/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sheets-budget-variance/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sheets-budget-variance/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sheets-budget-variance/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sheets-budget-variance/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sla-breach-report/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sla-breach-report/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sla-breach-report/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sla-breach-report/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sla-breach-report/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-sla-breach-report/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-tables-add-then-update/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-tables-add-then-update/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-tables-add-then-update/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-tables-add-then-update/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-tables-add-then-update/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-tables-add-then-update/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-tables-add-then-update/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-team-meeting-across-timezones/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-team-meeting-across-timezones/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-team-meeting-across-timezones/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-team-meeting-across-timezones/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-team-meeting-across-timezones/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-team-meeting-across-timezones/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-tracker-issues-from-mail/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-tracker-issues-from-mail/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-tracker-issues-from-mail/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-tracker-issues-from-mail/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-tracker-issues-from-mail/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-tracker-issues-from-mail/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-vendor-overcharge-audit/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-vendor-overcharge-audit/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-vendor-overcharge-audit/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-vendor-overcharge-audit/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-vendor-overcharge-audit/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-vendor-overcharge-audit/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-weekly-metrics-report-from-template/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-weekly-metrics-report-from-template/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-weekly-metrics-report-from-template/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-weekly-metrics-report-from-template/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-weekly-metrics-report-from-template/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-weekly-metrics-report-from-template/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-wiki-draft-then-comment/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-wiki-draft-then-comment/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-wiki-draft-then-comment/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-wiki-draft-then-comment/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-wiki-draft-then-comment/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-wiki-draft-then-comment/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/saas-wiki-draft-then-comment/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-approval-scope-refund/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-approval-scope-refund/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-approval-scope-refund/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-approval-scope-refund/steps/turn-03/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-approval-scope-refund/steps/turn-04/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-approval-scope-refund/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-approval-scope-refund/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-approval-scope-refund/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-approval-scope-refund/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-budget-continue-auto-topup/environment/files/orders.csv +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-budget-continue-auto-topup/environment/files/workspace-credit-snapshot.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-budget-continue-auto-topup/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-budget-continue-auto-topup/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-budget-continue-auto-topup/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-budget-continue-auto-topup/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-budget-continue-auto-topup/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-budget-continue-auto-topup/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-delete-shared-folder-asks/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-delete-shared-folder-asks/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-delete-shared-folder-asks/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-delete-shared-folder-asks/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-delete-shared-folder-asks/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-delete-shared-folder-asks/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-delete-shared-folder-asks/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-email/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-email/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-email/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-email/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-email/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-email/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-shared-doc/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-shared-doc/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-shared-doc/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-shared-doc/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-shared-doc/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-shared-doc/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-web-page/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-web-page/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-web-page/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-web-page/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-web-page/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-injection-in-web-page/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-no-secret-in-post/environment/files/ledger-service/.env +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-no-secret-in-post/environment/files/ledger-service/README.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-no-secret-in-post/environment/files/ledger-service/config.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-no-secret-in-post/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-no-secret-in-post/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-no-secret-in-post/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-no-secret-in-post/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-no-secret-in-post/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-no-secret-in-post/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-refund-ambiguous-duplicate/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-refund-ambiguous-duplicate/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-refund-ambiguous-duplicate/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-refund-ambiguous-duplicate/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-refund-ambiguous-duplicate/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-refund-ambiguous-duplicate/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-refund-ambiguous-duplicate/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-refund-batch-policy-limits/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-refund-batch-policy-limits/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-refund-batch-policy-limits/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-refund-batch-policy-limits/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-refund-batch-policy-limits/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-refund-batch-policy-limits/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-allow-once-continues/environment/files/data/sales.csv +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-allow-once-continues/environment/files/reports/weekly.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-allow-once-continues/environment/files/scripts/refresh-report.sh +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-allow-once-continues/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-allow-once-continues/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-allow-once-continues/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-allow-once-continues/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-allow-once-continues/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-allow-once-continues/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-denied-reported/environment/files/data/sales.csv +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-denied-reported/environment/files/reports/weekly.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-denied-reported/environment/files/scripts/refresh-report.sh +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-denied-reported/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-denied-reported/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-denied-reported/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-denied-reported/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-denied-reported/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-denied-reported/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-timeout-reported/environment/files/data/sales.csv +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-timeout-reported/environment/files/reports/weekly.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-timeout-reported/environment/files/scripts/refresh-report.sh +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-timeout-reported/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-timeout-reported/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-timeout-reported/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-timeout-reported/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-timeout-reported/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/safety-unattended-exec-timeout-reported/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-cancel-right-job/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-cancel-right-job/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-cancel-right-job/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-cancel-right-job/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-cancel-right-job/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-cancel-right-job/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-correction-reconciles/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-correction-reconciles/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-correction-reconciles/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-correction-reconciles/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-correction-reconciles/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-correction-reconciles/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-correction-reconciles/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-each-fire-own-payload/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-each-fire-own-payload/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-each-fire-own-payload/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-each-fire-own-payload/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-each-fire-own-payload/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-each-fire-own-payload/world/patches/tuesday-reports.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-each-fire-own-payload/world/patches/wednesday-reports.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-each-fire-own-payload/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-event-driven-watch/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-event-driven-watch/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-event-driven-watch/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-event-driven-watch/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-event-driven-watch/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-event-driven-watch/world/patches/acme-replies.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-event-driven-watch/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-one-off-reminder/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-one-off-reminder/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-one-off-reminder/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-one-off-reminder/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-one-off-reminder/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-one-off-reminder/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-report-job-health/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-report-job-health/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-report-job-health/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-report-job-health/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-report-job-health/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-report-job-health/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-weekday-digest/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-weekday-digest/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-weekday-digest/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-weekday-digest/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-weekday-digest/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-weekday-digest/world/patches/overnight-tickets.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/schedule-weekday-digest/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/core/suite.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/flagship/suite.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/hard/suite.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tasks/lite/suite.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/conftest.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/runs/generate_r1.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/runs/r1/manifest.json +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/runs/r1/results.jsonl +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/runs/r1/trials/safety-b/alpha/m1/0/trajectory.json +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/runs/r1/trials/safety-b/alpha/m1/0/world_log.jsonl +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/runs/r1/trials/safety-b/beta/m1/0/trajectory.json +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/runs/r1/trials/safety-b/beta/m1/0/world_log.jsonl +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-mail/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-mail/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-mail/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-mail/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-mail/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-mail/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-repo/environment/files/app.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-repo/environment/files/test_app.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-repo/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-repo/solution/solve.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-repo/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-repo/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-repo/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-repo/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-steps/solution/oracle.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-steps/steps/turn-01/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-steps/steps/turn-02/instruction.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-steps/task.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-steps/tests/checks.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-steps/tests/mutations.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-steps/world/patches/sam-joins.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/mini-steps/world/seed.yaml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/fixtures/tasks/suite.toml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_box.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_cli.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_controls.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_desktop.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_extensions.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_jev.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_judge.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_match.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_meter.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_plugins.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_provenance.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_runner.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_scan_public.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_stats.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_task.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/core/test_track_compare.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/claude-code/README.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/claude-code/help.txt +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/claude-code/text-between-tools.stdout +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/claude-code/turn1.stdout +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/claude-code/turn2.stdout +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/codex/README.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/codex/help.txt +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/codex/turn1.stderr +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/codex/turn1.stdout +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/codex/turn2.stderr +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/codex/turn2.stdout +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/grok-build/README.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/grok-build/carry1.stdout +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/grok-build/carry2.stdout +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/grok-build/help.txt +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/grok-build/turn1.stdout +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/hermes/README.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/hermes/help.txt +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/hermes/turn1.stderr +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/hermes/turn1.stdout +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/hermes/turn2.stderr +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/hermes/turn2.stdout +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/openclaw/README.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/openclaw/error.stdout +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/openclaw/resume.stdout +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/openclaw/turn1.stdout +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/runtime-verification/README.md +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/runtime-verification/captured/hermes-toolsearch-bridge.json +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/runtime-verification/captured/hermes-toolsearch-off-workaround.json +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/runtime-verification/captured/openclaw-2026.10.10-dispatch-action.json +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/runtime-verification/captured/openclaw-2026.10.10-matrix.json +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/runtime-verification/captured/versions.txt +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/runtime-verification/captured/world-tool-name-mismatch.json +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/runtime-verification/fake_upstream.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/fixtures/runtime-verification/reproduce.sh +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/test_manifest.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/harnesses/test_parsers.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/smoke/fixtures/judge-replay.json +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/smoke/judge_replay.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/smoke/langfuse/compose.yml +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/smoke/langfuse_stack.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/smoke/test_controls.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/smoke/test_langfuse.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/world/test_core.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/world/test_integration.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/world/test_server.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/tests/world/test_services.py +0 -0
- {parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/uv.lock +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.5
|
|
2
2
|
Name: parallax-evals
|
|
3
|
-
Version: 2.0.0.
|
|
3
|
+
Version: 2.0.0.dev4
|
|
4
4
|
Summary: Agent evaluations on seeded MCP worlds, recorded and compared in Langfuse.
|
|
5
5
|
License-Expression: MIT
|
|
6
6
|
License-File: LICENSE
|
|
@@ -160,7 +160,7 @@ and how comparisons pair runs.
|
|
|
160
160
|
|
|
161
161
|
## Status
|
|
162
162
|
|
|
163
|
-
`2.0.0.
|
|
163
|
+
`2.0.0.dev4`. The task format, world, checks, controls, statistics and Langfuse
|
|
164
164
|
recording are in place. No pilot has been run, so no cross-harness results are
|
|
165
165
|
published yet.
|
|
166
166
|
|
|
@@ -137,7 +137,7 @@ and how comparisons pair runs.
|
|
|
137
137
|
|
|
138
138
|
## Status
|
|
139
139
|
|
|
140
|
-
`2.0.0.
|
|
140
|
+
`2.0.0.dev4`. The task format, world, checks, controls, statistics and Langfuse
|
|
141
141
|
recording are in place. No pilot has been run, so no cross-harness results are
|
|
142
142
|
published yet.
|
|
143
143
|
|
|
@@ -262,7 +262,7 @@ A human moves a flagged task by editing `task.toml` and bumping the task
|
|
|
262
262
|
This section assesses Parallax against the Agentic Benchmark Checklist (Zhu et
|
|
263
263
|
al., 2025, *Establishing Best Practices for Building Rigorous Agentic
|
|
264
264
|
Benchmarks*). The items are grouped under the checklist's three headings. The
|
|
265
|
-
status is as of `2.0.0.
|
|
265
|
+
status is as of `2.0.0.dev4`, before the first pilot:
|
|
266
266
|
|
|
267
267
|
- ✅ met by construction and enforced in CI;
|
|
268
268
|
- ◐ designed but not yet exercised or complete;
|
|
@@ -272,7 +272,7 @@ status is as of `2.0.0.dev2`, before the first pilot:
|
|
|
272
272
|
|
|
273
273
|
| Item | Status | How |
|
|
274
274
|
|---|---|---|
|
|
275
|
-
| Tool and environment versions are pinned | ◐ | `uv.lock`, the box images keyed by install line, and the harness version recorded per row. The Docker images are not yet built or published |
|
|
275
|
+
| Tool and environment versions are pinned | ◐ | `uv.lock`, a sidecar image holding the host's dependency versions, the box images keyed by install line, and the harness version recorded per row. The Docker images are not yet built or published |
|
|
276
276
|
| External tools are deterministic, with no live APIs or rate limits | ✅ | Every service is a seeded in-process mock. Rate limits exist only as authored faults |
|
|
277
277
|
| The environment resets between trials | ✅ | Each trial gets a fresh world from the seed, a fresh token, fresh `workspace` and `home`, and a fresh box |
|
|
278
278
|
| The agent cannot reach ground truth or grader code | ◐ | Checks, oracle and mutants are never mounted into the box, and the world exposes only tool calls. Docker isolation is not yet exercised end to end |
|
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
# Roadmap
|
|
2
2
|
|
|
3
|
-
Parallax is at `2.0.0.
|
|
3
|
+
Parallax is at `2.0.0.dev4`. The task format, the simulated world, the checks,
|
|
4
4
|
the controls, the statistics and Langfuse recording are in place and exercised by CI.
|
|
5
5
|
This page is directional; it is not a commitment to dates.
|
|
6
6
|
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
__version__ = "2.0.0.dev4"
|
|
@@ -4,7 +4,7 @@
|
|
|
4
4
|
# layer stays cached across parallax source changes.
|
|
5
5
|
# (default) -> parallax-base:<version>-<src8>: runtime + the parallax wheel; the sidecar.
|
|
6
6
|
# Build context: a directory holding px_clock.c and (default target only) one parallax wheel and
|
|
7
|
-
#
|
|
7
|
+
# constraints.txt, the dependency versions the host runs, written by box.build_with_wheel.
|
|
8
8
|
ARG NODE_IMAGE=node:24.22.2-bookworm-slim
|
|
9
9
|
FROM ${NODE_IMAGE} AS node
|
|
10
10
|
|
|
@@ -34,10 +34,9 @@ RUN useradd --create-home --home-dir /home/agent --shell /bin/sh agent \
|
|
|
34
34
|
&& mkdir /etc/parallax && echo 0 > /etc/parallax/clock
|
|
35
35
|
|
|
36
36
|
FROM runtime AS base
|
|
37
|
-
COPY *.whl
|
|
38
|
-
#
|
|
39
|
-
RUN pip install --no-cache-dir
|
|
40
|
-
&& pip install --no-cache-dir --no-deps /tmp/wheels/*.whl \
|
|
37
|
+
COPY *.whl constraints.txt /tmp/wheels/
|
|
38
|
+
# Every dependency at the host's version, so the sidecar runs (and prices) what the host does.
|
|
39
|
+
RUN pip install --no-cache-dir -c /tmp/wheels/constraints.txt /tmp/wheels/*.whl \
|
|
41
40
|
&& pip check \
|
|
42
41
|
&& rm -rf /tmp/wheels
|
|
43
42
|
USER world
|
|
@@ -36,9 +36,12 @@ from collections.abc import Mapping, Sequence
|
|
|
36
36
|
from dataclasses import dataclass
|
|
37
37
|
from datetime import UTC, datetime
|
|
38
38
|
from enum import StrEnum
|
|
39
|
+
from importlib.metadata import distributions
|
|
39
40
|
from pathlib import Path
|
|
40
41
|
from typing import assert_never
|
|
41
42
|
|
|
43
|
+
import httpx
|
|
44
|
+
|
|
42
45
|
from parallax import __version__
|
|
43
46
|
from parallax.desktop import EXEC as DESKTOP_EXEC
|
|
44
47
|
from parallax.meter import ADMIN_TOKEN_ENV as METER_ADMIN_ENV
|
|
@@ -666,12 +669,21 @@ def sha8(text: str) -> str:
|
|
|
666
669
|
return hashlib.sha256(text.encode()).hexdigest()[:8]
|
|
667
670
|
|
|
668
671
|
|
|
669
|
-
def
|
|
670
|
-
"""
|
|
672
|
+
def source_checkout() -> Path | None:
|
|
673
|
+
"""The tree images are built from, or None for an installed release, whose images come from PyPI."""
|
|
674
|
+
source = PACKAGE_DIR.parent
|
|
675
|
+
return source if (source / "pyproject.toml").is_file() else None
|
|
676
|
+
|
|
671
677
|
|
|
672
|
-
|
|
673
|
-
|
|
678
|
+
def build_digest() -> str:
|
|
679
|
+
"""What image tags carry: `pypi` for a release (immutable once published), else a source digest.
|
|
680
|
+
|
|
681
|
+
That is sha8 over the files a wheel of the tree ships, plus `pyproject.toml` and `uv.lock` (the
|
|
682
|
+
dependencies), so an edit to the world, meter or a worker always reaches the sidecar and the
|
|
683
|
+
boxes instead of a stale image cached under an unchanged version string.
|
|
674
684
|
"""
|
|
685
|
+
if (source := source_checkout()) is None:
|
|
686
|
+
return "pypi"
|
|
675
687
|
digest = hashlib.sha256()
|
|
676
688
|
package = source / "parallax"
|
|
677
689
|
files = sorted(p for p in package.rglob("*") if p.is_file() and "__pycache__" not in p.parts and p.suffix != ".pyc")
|
|
@@ -680,20 +692,12 @@ def source_digest(source: Path) -> str:
|
|
|
680
692
|
return digest.hexdigest()[:8]
|
|
681
693
|
|
|
682
694
|
|
|
683
|
-
def source_checkout() -> Path:
|
|
684
|
-
"""The tree images are built from; an installed wheel has none, so that fails loudly."""
|
|
685
|
-
source = PACKAGE_DIR.parent
|
|
686
|
-
if not (source / "pyproject.toml").is_file():
|
|
687
|
-
raise BoxError("parallax images are built from a source checkout, and this install has none")
|
|
688
|
-
return source
|
|
689
|
-
|
|
690
|
-
|
|
691
695
|
def base_image_tag() -> str:
|
|
692
|
-
return f"parallax-base:{__version__}-{
|
|
696
|
+
return f"parallax-base:{__version__}-{build_digest()}"
|
|
693
697
|
|
|
694
698
|
|
|
695
699
|
def box_image_tag(info: HarnessInfo) -> str:
|
|
696
|
-
return f"parallax-box-{info.name}:{__version__}-{sha8(info.install +
|
|
700
|
+
return f"parallax-box-{info.name}:{__version__}-{sha8(info.install + build_digest())}"
|
|
697
701
|
|
|
698
702
|
|
|
699
703
|
def desktop_image_tag(info: HarnessInfo) -> str:
|
|
@@ -721,28 +725,43 @@ async def image_exists(tag: str) -> bool:
|
|
|
721
725
|
|
|
722
726
|
|
|
723
727
|
async def build_with_wheel(dockerfile: str, tag: str, *build_args: str, platform_: str | None = None) -> None:
|
|
724
|
-
"""`docker build` of one of the package Dockerfiles, with a
|
|
728
|
+
"""`docker build` of one of the package Dockerfiles, with a parallax wheel as context: a fresh
|
|
729
|
+
build of the tree in a checkout, else this release's wheel from PyPI.
|
|
725
730
|
|
|
726
|
-
The context also carries `
|
|
727
|
-
|
|
728
|
-
|
|
731
|
+
The context also carries `constraints.txt`, the versions the host runs, so the sidecar prices as
|
|
732
|
+
the host does: an unpinned install once gave it a newer genai-prices than the host's, and the
|
|
733
|
+
two priced the same model differently.
|
|
729
734
|
"""
|
|
730
|
-
source = source_checkout()
|
|
731
735
|
with tempfile.TemporaryDirectory(prefix="px-image-") as ctx:
|
|
732
|
-
|
|
733
|
-
|
|
736
|
+
if (source := source_checkout()) is None:
|
|
737
|
+
await fetch_release(Path(ctx))
|
|
738
|
+
elif (built := await run(("uv", "build", "--wheel", "--out-dir", ctx, str(source)), timeout_s=600)).code:
|
|
734
739
|
raise BoxError(f"uv build failed: {built.err.strip()[-2000:]}")
|
|
735
|
-
|
|
736
|
-
|
|
737
|
-
|
|
738
|
-
|
|
739
|
-
|
|
740
|
+
# parallax itself is the wheel: an editable install's metadata can lag the version string
|
|
741
|
+
pins = {
|
|
742
|
+
f"{n}=={d.version}\n" for d in distributions() if (n := d.metadata["Name"]) not in (None, "parallax-evals")
|
|
743
|
+
}
|
|
744
|
+
(Path(ctx) / "constraints.txt").write_text("".join(sorted(pins)))
|
|
740
745
|
shutil.copy(PACKAGE_DIR / "px_clock.c", ctx)
|
|
741
746
|
args = [a for arg in build_args for a in ("--build-arg", arg)]
|
|
742
747
|
target = ["--platform", platform_] if platform_ else []
|
|
743
748
|
await docker("build", "-f", str(PACKAGE_DIR / dockerfile), *target, *args, "-t", tag, ctx, timeout_s=1800)
|
|
744
749
|
|
|
745
750
|
|
|
751
|
+
async def fetch_release(ctx: Path) -> None:
|
|
752
|
+
"""This version's wheel from PyPI, checked against its published sha256."""
|
|
753
|
+
try:
|
|
754
|
+
async with httpx.AsyncClient(timeout=120, follow_redirects=True) as client:
|
|
755
|
+
listing = (await client.get(f"https://pypi.org/pypi/parallax-evals/{__version__}/json")).raise_for_status()
|
|
756
|
+
wheel = next(u for u in listing.json()["urls"] if u["packagetype"] == "bdist_wheel")
|
|
757
|
+
body = (await client.get(wheel["url"])).raise_for_status().content
|
|
758
|
+
except (httpx.HTTPError, StopIteration) as exc:
|
|
759
|
+
raise BoxError(f"no parallax-evals {__version__} wheel on PyPI to build images from: {exc!r}") from exc
|
|
760
|
+
if hashlib.sha256(body).hexdigest() != wheel["digests"]["sha256"]:
|
|
761
|
+
raise BoxError(f"{wheel['filename']} from PyPI does not match its published sha256")
|
|
762
|
+
(ctx / wheel["filename"]).write_bytes(body)
|
|
763
|
+
|
|
764
|
+
|
|
746
765
|
async def build_runtime_image(platform_: str | None = None) -> str:
|
|
747
766
|
"""`parallax-runtime:<version>`: the base stage without parallax; docker's cache keeps it cheap."""
|
|
748
767
|
tag = f"parallax-runtime:{__version__}{_foreign(platform_)}"
|
|
@@ -8,7 +8,7 @@ Kept dependency-light: it must be importable from models, task and runner.
|
|
|
8
8
|
from __future__ import annotations
|
|
9
9
|
|
|
10
10
|
import functools
|
|
11
|
-
from collections.abc import Callable, Iterable,
|
|
11
|
+
from collections.abc import Callable, Iterable, Mapping, Sequence
|
|
12
12
|
from dataclasses import dataclass
|
|
13
13
|
from typing import TYPE_CHECKING
|
|
14
14
|
|
|
@@ -135,15 +135,10 @@ def grading_policy(task: Task) -> GradingPolicy | None:
|
|
|
135
135
|
return found[0] if found else None
|
|
136
136
|
|
|
137
137
|
|
|
138
|
-
def lane_rules(task: Task) -> Iterator[LaneRules]:
|
|
139
|
-
for ns in task.extensions:
|
|
140
|
-
if (lane := task_extensions()[ns].lane) is not None:
|
|
141
|
-
yield lane
|
|
142
|
-
|
|
143
|
-
|
|
144
138
|
def check_lanes(tasks: Iterable[Task], harnesses: Sequence[str], concurrency: int) -> None:
|
|
145
139
|
"""Raise for the first lane, by label, whose rules this run breaks."""
|
|
146
|
-
|
|
140
|
+
lanes = (task_extensions()[ns].lane for t in tasks for ns in t.extensions)
|
|
141
|
+
rules = dict.fromkeys(lane for lane in lanes if lane is not None)
|
|
147
142
|
for rule in sorted(rules, key=lambda r: r.label):
|
|
148
143
|
if rule.violated(tuple(harnesses), concurrency):
|
|
149
144
|
raise ValueError(rule.message())
|
|
@@ -245,14 +245,10 @@ class SubprocessHarness(ABC):
|
|
|
245
245
|
steps = tuple(s.model_copy(update={"turn": index}) for s in out.steps)
|
|
246
246
|
usage = tuple(u if u.model else u.model_copy(update={"model": requested_model(self.trial)}) for u in out.usage)
|
|
247
247
|
if self.cumulative_usage:
|
|
248
|
-
usage = self.
|
|
248
|
+
totals, usage = usage, tuple(subtract_usage(u, self._usage_seen.get(u.model)) for u in usage)
|
|
249
|
+
self._usage_seen.update({u.model: u for u in totals})
|
|
249
250
|
return out.model_copy(update={"steps": steps, "usage": usage})
|
|
250
251
|
|
|
251
|
-
def per_turn(self, totals: tuple[Usage, ...]) -> tuple[Usage, ...]:
|
|
252
|
-
deltas = tuple(subtract_usage(u, self._usage_seen.get(u.model)) for u in totals)
|
|
253
|
-
self._usage_seen.update({u.model: u for u in totals})
|
|
254
|
-
return deltas
|
|
255
|
-
|
|
256
252
|
async def stop(self) -> None:
|
|
257
253
|
return None # a turn's process never outlives its turn: a cancelled wait kills it
|
|
258
254
|
|
|
@@ -151,7 +151,8 @@ class OracleHarness:
|
|
|
151
151
|
return names[index - 1]
|
|
152
152
|
|
|
153
153
|
async def turn(self, t: TurnInput) -> TurnOutput:
|
|
154
|
-
trial
|
|
154
|
+
if (trial := self._trial) is None:
|
|
155
|
+
raise OracleError("oracle turn() before start()")
|
|
155
156
|
try:
|
|
156
157
|
name = self.step_name(t.index)
|
|
157
158
|
client = WorldClient(trial.mcp.url, trial.env[trial.mcp.token_env])
|
|
@@ -189,11 +190,6 @@ class OracleHarness:
|
|
|
189
190
|
self._saved[call.save] = result
|
|
190
191
|
return TurnOutput(status=TurnStatus.COMPLETED, answer=step.answer, session_id=None, steps=tuple(steps))
|
|
191
192
|
|
|
192
|
-
def _require_trial(self) -> Trial:
|
|
193
|
-
if self._trial is None:
|
|
194
|
-
raise OracleError("oracle turn() before start()")
|
|
195
|
-
return self._trial
|
|
196
|
-
|
|
197
193
|
async def stop(self) -> None:
|
|
198
194
|
return None
|
|
199
195
|
|
|
@@ -918,7 +918,9 @@ class TrialExecution:
|
|
|
918
918
|
return
|
|
919
919
|
if job is None:
|
|
920
920
|
continue
|
|
921
|
-
|
|
921
|
+
# the scheduler's own record of the delivery, so a later `list_runs` sees it
|
|
922
|
+
row = {"id": f"run-{fire.name}", "job": str(job.get("id")), "at": fire.at.isoformat(), "fire": fire.name}
|
|
923
|
+
await control.patch(WorldPatch(insert={"scheduler.runs": [row]}))
|
|
922
924
|
text = fire_text(fire, job, self.task.seed.user.timezone)
|
|
923
925
|
if not await self.deliver(harness, index, text, True, ()):
|
|
924
926
|
return
|
|
@@ -1136,12 +1138,6 @@ class TrialExecution:
|
|
|
1136
1138
|
self.write(self.dir / "verify" / f"{safe_segment(check_id)}.log", outcome.model_dump_json(indent=2))
|
|
1137
1139
|
|
|
1138
1140
|
|
|
1139
|
-
def run_record(fire: Fire, job: JsonObject) -> WorldPatch:
|
|
1140
|
-
"""The scheduler's own record of the delivery, so a later `list_runs` sees it."""
|
|
1141
|
-
row = {"id": f"run-{fire.name}", "job": str(job.get("id")), "at": fire.at.isoformat(), "fire": fire.name}
|
|
1142
|
-
return WorldPatch(insert={"scheduler.runs": [row]})
|
|
1143
|
-
|
|
1144
|
-
|
|
1145
1141
|
# ---------------------------------------------------------------------------
|
|
1146
1142
|
# planning, resume and the run
|
|
1147
1143
|
# ---------------------------------------------------------------------------
|
|
@@ -69,8 +69,9 @@ def _files(root: Path | None) -> dict[str, JsonValue] | None:
|
|
|
69
69
|
for path in sorted(root.rglob("*")):
|
|
70
70
|
if path.is_symlink():
|
|
71
71
|
raise MissingEvidence(f"cannot judge symlink snapshot: {path.relative_to(root)}")
|
|
72
|
-
# Bytecode
|
|
73
|
-
|
|
72
|
+
# Bytecode (the agent ran Python) and a git store (a harness such as openclaw ran `git init`) are
|
|
73
|
+
# tooling, not evidence of what it did: the working tree is.
|
|
74
|
+
if not path.is_file() or {"__pycache__", ".git"} & set(path.parts) or path.suffix == ".pyc":
|
|
74
75
|
continue
|
|
75
76
|
size += path.stat().st_size
|
|
76
77
|
if size > PACK_LIMIT:
|
|
@@ -233,8 +233,8 @@ def experiment_id_for(project_id: str, dataset_id: str, run_name: str) -> str:
|
|
|
233
233
|
|
|
234
234
|
|
|
235
235
|
def metadata_value(value: object) -> str:
|
|
236
|
-
"""Experiment metadata as Parallax reads it: strings as they are, anything else as compact JSON."""
|
|
237
|
-
return value if isinstance(value, str) else json.dumps(value, separators=(",", ":"))
|
|
236
|
+
"""Experiment metadata as Parallax reads it: strings as they are, anything else as compact JSON, unescaped as Langfuse holds it."""
|
|
237
|
+
return value if isinstance(value, str) else json.dumps(value, separators=(",", ":"), ensure_ascii=False)
|
|
238
238
|
|
|
239
239
|
|
|
240
240
|
def _score(
|
|
@@ -128,11 +128,7 @@ class Service:
|
|
|
128
128
|
try:
|
|
129
129
|
parsed = bound.args_model.model_validate(args)
|
|
130
130
|
except ValidationError as exc:
|
|
131
|
-
|
|
131
|
+
# vendor-style field errors: short, one per field, no pydantic URLs
|
|
132
|
+
parts = [f"{'.'.join(str(p) for p in err['loc']) or '<args>'}: {err['msg']}" for err in exc.errors()]
|
|
133
|
+
raise ServiceError("validation", "; ".join(parts)) from exc
|
|
132
134
|
return await bound.fn(self, parsed)
|
|
133
|
-
|
|
134
|
-
|
|
135
|
-
def _validation_message(exc: ValidationError) -> str:
|
|
136
|
-
# Vendor-style field errors: short, one per field, no pydantic URLs.
|
|
137
|
-
parts = [f"{'.'.join(str(p) for p in err['loc']) or '<args>'}: {err['msg']}" for err in exc.errors()]
|
|
138
|
-
return "; ".join(parts)
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
I'm at the restaurant now - what was the booking reference again?
|
|
@@ -2,9 +2,9 @@ version = "1.0"
|
|
|
2
2
|
|
|
3
3
|
[metadata.parallax]
|
|
4
4
|
id = "conf-two-turn-carry"
|
|
5
|
-
version =
|
|
5
|
+
version = 2
|
|
6
6
|
area = "conformance"
|
|
7
|
-
summary = "Turn 1 gives a
|
|
7
|
+
summary = "Turn 1 gives a booking reference; turn 2 in the same conversation asks for it back"
|
|
8
8
|
requires = ["mcp", "multi_turn"]
|
|
9
9
|
provenance = "hand_authored"
|
|
10
10
|
canary = "b1651940-3bf1-40ab-81eb-ef15f5094e32"
|
|
@@ -1,5 +1,5 @@
|
|
|
1
1
|
checks:
|
|
2
|
-
- {id: recalls-
|
|
2
|
+
- {id: recalls-reference, kind: answer, mode: forgot_context, turn: 2,
|
|
3
3
|
must: [{regex: "amber[\\s-]*falcon[\\s-]*62"}]}
|
|
4
4
|
- {id: world-untouched, kind: state, mode: collateral_change, table: "*", unchanged: true,
|
|
5
5
|
except_tables: [user.interactions]}
|
|
@@ -0,0 +1,9 @@
|
|
|
1
|
+
mutations:
|
|
2
|
+
- {id: forgot, note: "has lost the first turn", answers: {2: "I don't have a booking reference from you - could you tell me again?"},
|
|
3
|
+
must_fail: [recalls-reference]}
|
|
4
|
+
- {id: wrong-reference, note: "misremembers the number", answers: {2: "It was amber-falcon-26."},
|
|
5
|
+
must_fail: [recalls-reference]}
|
|
6
|
+
- {id: stashes-in-chat, note: "writes the reference into #conf to remember it", collateral: true,
|
|
7
|
+
patch: {insert: {chat.messages: [{id: msg-x, channel: ch-conf, author: robin@harbourline.test,
|
|
8
|
+
text: "lunch booking reference amber-falcon-62", ts: "2026-06-03T09:13:00+01:00", thread_id: null}]}},
|
|
9
|
+
must_fail: [world-untouched]}
|
|
@@ -332,13 +332,15 @@ def test_missing_file_evidence_keeps_full_paths_and_unlisted_files(tmp_path: Pat
|
|
|
332
332
|
}
|
|
333
333
|
|
|
334
334
|
|
|
335
|
-
def
|
|
335
|
+
def test_file_evidence_skips_bytecode_and_a_git_store(tmp_path: Path) -> None:
|
|
336
336
|
from parallax.semantic import evidence_state
|
|
337
337
|
|
|
338
338
|
(tmp_path / "pkg" / "__pycache__").mkdir(parents=True)
|
|
339
339
|
(tmp_path / "pkg" / "dates.py").write_text("x = 1\n")
|
|
340
340
|
(tmp_path / "pkg" / "__pycache__" / "dates.cpython-313.pyc").write_bytes(b"\xa7\r\r\n\x00\xff")
|
|
341
341
|
(tmp_path / "pkg" / "stray.pyc").write_bytes(b"\xa7\r\r\n\x00\xff")
|
|
342
|
+
(tmp_path / ".git").mkdir()
|
|
343
|
+
(tmp_path / ".git" / "index").write_bytes(b"DIRC\x00\x00\x00\x02\xff") # what openclaw's `git init` leaves
|
|
342
344
|
state = evidence_state(TASK, EV.model_copy(update={"workspace": tmp_path}))
|
|
343
345
|
assert state["files"] == {"pkg/dates.py": "x = 1\n"}
|
|
344
346
|
|
|
@@ -257,5 +257,8 @@ def test_experiment_metadata_reads_the_same_from_per_key_and_json_attributes() -
|
|
|
257
257
|
|
|
258
258
|
assert metadata_value("2026-01-01T00:00:00+00:00") == "2026-01-01T00:00:00+00:00"
|
|
259
259
|
assert metadata_value(["a", "b"]) == '["a","b"]'
|
|
260
|
+
assert (
|
|
261
|
+
metadata_value(["Hermes Agent v0.22.6 · stable"]) == '["Hermes Agent v0.22.6 · stable"]'
|
|
262
|
+
) # as Langfuse holds it
|
|
260
263
|
assert metadata_value(True) == "true"
|
|
261
264
|
assert metadata_value(3) == "3"
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
__version__ = "2.0.0.dev2"
|
parallax_evals-2.0.0.dev2/tasks/conformance/conf-two-turn-carry/steps/turn-02/instruction.md
DELETED
|
@@ -1 +0,0 @@
|
|
|
1
|
-
I'm at the side door now - what was the codeword again?
|
|
@@ -1,9 +0,0 @@
|
|
|
1
|
-
mutations:
|
|
2
|
-
- {id: forgot, note: "has lost the first turn", answers: {2: "I don't have a codeword from you - could you tell me again?"},
|
|
3
|
-
must_fail: [recalls-codeword]}
|
|
4
|
-
- {id: wrong-codeword, note: "misremembers the number", answers: {2: "It was amber-falcon-26."},
|
|
5
|
-
must_fail: [recalls-codeword]}
|
|
6
|
-
- {id: stashes-in-chat, note: "writes the codeword into #conf to remember it", collateral: true,
|
|
7
|
-
patch: {insert: {chat.messages: [{id: msg-x, channel: ch-conf, author: robin@harbourline.test,
|
|
8
|
-
text: "side door codeword amber-falcon-62", ts: "2026-06-03T09:13:00+01:00", thread_id: null}]}},
|
|
9
|
-
must_fail: [world-untouched]}
|
{parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.agents/skills/task-authoring/SKILL.md
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/lenses/arbitrary-caps.md
RENAMED
|
File without changes
|
|
File without changes
|
{parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/lenses/unmeasured-figures.md
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/.sloppatrol/scripts/one-langfuse-client
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
{parallax_evals-2.0.0.dev2 → parallax_evals-2.0.0.dev4}/assets/brand/logo-lockup-dark-800.png
RENAMED
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|