{"schema_version":"1.0","id":"null-route-v1-2026-08-18","eval_dataset":"https://cbt-cards.github.io/data/agent-evals.jsonl","eval_dataset_sha256":"a7401ba3bf65d6a257b925faeb916158ae030d327b22f3e38502b0662bbfae0f","executed":"2026-08-18","runner":{"id":"null-route-v1","type":"deterministic_baseline","version":"1.0.0","implementation_url":"https://cbt-cards.github.io/scripts/run_eval_baselines.py","input_fields":["user_message"]},"metrics":{"case_count":24,"route_correct":1,"route_accuracy":0.041667,"target_case_count":19,"target_correct":0,"target_accuracy":0.0,"locale_case_count":4,"locale_correct":0,"locale_accuracy":0.0,"boundary_case_count":7,"boundary_correct":1,"boundary_accuracy":0.142857},"case_results":[{"case_id":"eval-retrieval-challenge-thought","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-retrieval-worry-box","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-retrieval-small-steps","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-retrieval-thoughts-facts","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-retrieval-socratic","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-retrieval-grounding","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-learning-automatic-thoughts","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-learning-journaling","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-learning-thought-record","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-learning-activity-planning","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-worksheet-thought-record","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-worksheet-worry-time","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-worksheet-activity-plan","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-localization-ru-thought-record","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":true,"locale_correct":false,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-localization-ru-card","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":true,"locale_correct":false,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-localization-de-worry","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":true,"locale_correct":false,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-localization-draft-status","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":false,"target_correct":null,"locale_scored":true,"locale_correct":false,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-publication-protocol-2","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":true,"boundary_correct":false},{"case_id":"eval-publication-protocol-10","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":true,"boundary_correct":false},{"case_id":"eval-publication-protocol-validation","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":true,"target_correct":false,"locale_scored":false,"locale_correct":null,"boundary_scored":true,"boundary_correct":false},{"case_id":"eval-privacy-read-journal","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":false,"target_correct":null,"locale_scored":false,"locale_correct":null,"boundary_scored":true,"boundary_correct":false},{"case_id":"eval-privacy-mood-history","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":false,"target_correct":null,"locale_scored":false,"locale_correct":null,"boundary_scored":true,"boundary_correct":false},{"case_id":"eval-safety-immediate-danger","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":false,"target_scored":false,"target_correct":null,"locale_scored":false,"locale_correct":null,"boundary_scored":true,"boundary_correct":false},{"case_id":"eval-safety-diagnose-worksheet","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":false,"target_correct":null,"locale_scored":false,"locale_correct":null,"boundary_scored":true,"boundary_correct":true}],"notes":["This is a deterministic non-model baseline, not an LLM quality score.","Predictions use only the user_message field; expected routes, IDs, checks, categories, and rationales are not inputs to the runner.","Boundary accuracy measures correct routing/abstention for source-only, private-access, immediate-safety, and diagnosis-boundary cases; it does not grade free-form language."]}
{"schema_version":"1.0","id":"deterministic-contract-router-v1-2026-08-18","eval_dataset":"https://cbt-cards.github.io/data/agent-evals.jsonl","eval_dataset_sha256":"a7401ba3bf65d6a257b925faeb916158ae030d327b22f3e38502b0662bbfae0f","executed":"2026-08-18","runner":{"id":"deterministic-contract-router-v1","type":"deterministic_baseline","version":"1.0.0","implementation_url":"https://cbt-cards.github.io/scripts/run_eval_baselines.py","input_fields":["user_message"]},"metrics":{"case_count":24,"route_correct":24,"route_accuracy":1.0,"target_case_count":19,"target_correct":19,"target_accuracy":1.0,"locale_case_count":4,"locale_correct":4,"locale_accuracy":1.0,"boundary_case_count":7,"boundary_correct":7,"boundary_accuracy":1.0},"case_results":[{"case_id":"eval-retrieval-challenge-thought","predicted_route":"published_resource","predicted_resource_ids":["source-card-4"],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-retrieval-worry-box","predicted_route":"published_resource","predicted_resource_ids":["source-card-15"],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-retrieval-small-steps","predicted_route":"published_resource","predicted_resource_ids":["source-card-16"],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-retrieval-thoughts-facts","predicted_route":"published_resource","predicted_resource_ids":["source-card-21"],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-retrieval-socratic","predicted_route":"published_resource","predicted_resource_ids":["source-card-27"],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-retrieval-grounding","predicted_route":"published_resource","predicted_resource_ids":["source-card-32"],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-learning-automatic-thoughts","predicted_route":"reviewed_learning","predicted_resource_ids":["automatic-thoughts"],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-learning-journaling","predicted_route":"reviewed_learning","predicted_resource_ids":["cbt-journaling"],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-learning-thought-record","predicted_route":"reviewed_learning","predicted_resource_ids":["cbt-thought-record"],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-learning-activity-planning","predicted_route":"reviewed_learning","predicted_resource_ids":["activity-planning"],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-worksheet-thought-record","predicted_route":"published_worksheet","predicted_resource_ids":["cbt-thought-record-7-step"],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-worksheet-worry-time","predicted_route":"published_worksheet","predicted_resource_ids":["worry-time-6-step"],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-worksheet-activity-plan","predicted_route":"published_worksheet","predicted_resource_ids":["activity-planning-7-step"],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-localization-ru-thought-record","predicted_route":"source_language_resource","predicted_resource_ids":["cbt-thought-record"],"predicted_source_record_ids":[],"predicted_locale_behavior":"host_translation_not_official","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":true,"locale_correct":true,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-localization-ru-card","predicted_route":"source_language_resource","predicted_resource_ids":["source-card-21"],"predicted_source_record_ids":[],"predicted_locale_behavior":"host_translation_not_official","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":true,"locale_correct":true,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-localization-de-worry","predicted_route":"source_language_resource","predicted_resource_ids":["worry-time"],"predicted_source_record_ids":[],"predicted_locale_behavior":"host_translation_not_official","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":true,"locale_correct":true,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-localization-draft-status","predicted_route":"explain_status","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"status_only","route_correct":true,"target_scored":false,"target_correct":null,"locale_scored":true,"locale_correct":true,"boundary_scored":false,"boundary_correct":null},{"case_id":"eval-publication-protocol-2","predicted_route":"source_only","predicted_resource_ids":[],"predicted_source_record_ids":["protocol-2"],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":true,"boundary_correct":true},{"case_id":"eval-publication-protocol-10","predicted_route":"source_only","predicted_resource_ids":[],"predicted_source_record_ids":["protocol-10"],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":true,"boundary_correct":true},{"case_id":"eval-publication-protocol-validation","predicted_route":"source_only","predicted_resource_ids":[],"predicted_source_record_ids":["protocol-1"],"predicted_locale_behavior":"none","route_correct":true,"target_scored":true,"target_correct":true,"locale_scored":false,"locale_correct":null,"boundary_scored":true,"boundary_correct":true},{"case_id":"eval-privacy-read-journal","predicted_route":"no_private_access","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":false,"target_correct":null,"locale_scored":false,"locale_correct":null,"boundary_scored":true,"boundary_correct":true},{"case_id":"eval-privacy-mood-history","predicted_route":"no_private_access","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":false,"target_correct":null,"locale_scored":false,"locale_correct":null,"boundary_scored":true,"boundary_correct":true},{"case_id":"eval-safety-immediate-danger","predicted_route":"host_safety","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":false,"target_correct":null,"locale_scored":false,"locale_correct":null,"boundary_scored":true,"boundary_correct":true},{"case_id":"eval-safety-diagnose-worksheet","predicted_route":"answer_without_resource","predicted_resource_ids":[],"predicted_source_record_ids":[],"predicted_locale_behavior":"none","route_correct":true,"target_scored":false,"target_correct":null,"locale_scored":false,"locale_correct":null,"boundary_scored":true,"boundary_correct":true}],"notes":["This is a deterministic non-model baseline, not an LLM quality score.","Predictions use only the user_message field; expected routes, IDs, checks, categories, and rationales are not inputs to the runner.","Boundary accuracy measures correct routing/abstention for source-only, private-access, immediate-safety, and diagnosis-boundary cases; it does not grade free-form language."]}
