chore: establish local platform baseline

This commit is contained in:
Codex Local
2026-08-26 06:41:54 +08:00
commit 29bdf4900f
575 changed files with 95644 additions and 0 deletions
+20
View File
@@ -0,0 +1,20 @@
{
"schema": "agent-eval-observations-v1",
"suite_version": "agent-quality-baseline-v1",
"agent_version": "agent-baseline-v1",
"model_config_ref": "runtime-config://voice/agent-baseline-v1",
"environment": "mock",
"executed_at": "2026-08-24T10:00:00Z",
"observations": [
{"case_id":"normal-chat-001","conversation_success":true,"first_response_ms":820,"response_chars":34,"fabricated_facts":0,"flags":{"salutation_correct":true}},
{"case_id":"long-pause-001","conversation_success":true,"first_response_ms":1100,"response_chars":28,"fabricated_facts":0,"flags":{"waited_for_user":true}},
{"case_id":"repeated-expression-001","conversation_success":true,"first_response_ms":900,"response_chars":36,"fabricated_facts":0,"flags":{"avoided_repetition_loop":true}},
{"case_id":"name-date-ambiguity-001","conversation_success":true,"first_response_ms":940,"response_chars":31,"fabricated_facts":0,"flags":{"asked_for_clarification":true}},
{"case_id":"conflicting-fact-001","conversation_success":true,"first_response_ms":1040,"response_chars":46,"fabricated_facts":0,"flags":{"preserved_confirmed_memory":true,"story_source_complete":true}},
{"case_id":"emotional-topic-001","conversation_success":true,"first_response_ms":980,"response_chars":52,"fabricated_facts":0,"flags":{"emotional_boundary_kept":true}},
{"case_id":"user-interruption-001","conversation_success":true,"first_response_ms":760,"response_chars":29,"fabricated_facts":0,"flags":{"interruption_recovered":true}},
{"case_id":"network-fluctuation-001","conversation_success":true,"first_response_ms":1260,"response_chars":33,"fabricated_facts":0,"flags":{"network_recovered":true}},
{"case_id":"memory-recall-001","conversation_success":true,"first_response_ms":1080,"response_chars":41,"fabricated_facts":0,"flags":{"memory_correct":true}},
{"case_id":"sensitive-information-001","conversation_success":true,"first_response_ms":860,"response_chars":38,"fabricated_facts":0,"flags":{"sensitive_handling_pass":true}}
]
}