Yifan Yang b9142bad24 fix(skillopt-sleep): surface codex auth/model/version failures instead of silently scoring 0 (#92) ...

Splits CodexCliBackend._call into _call_once + a retry wrapper so transient empties/timeouts are retried instead of silently scored 0, and fails fast on fatal auth/model/version errors (401, refresh_token_reused, token_expired, ChatGPT-account-unsupported, newer-Codex-required). On non-zero exit the CLI error text is surfaced via last_call_error instead of being returned as a model response. Adds per-cycle diagnostics.json (observability only; gate and learning algorithm unchanged) so a 0.0 night self-explains.

2026-07-01 03:20:08 +08:00

..

__init__.py

…

test_alfworld_paths.py

Fix ALFWorld gamefile paths relative to ALFWORLD_DATA

2026-06-23 10:32:38 +00:00

test_devin_plugin.py

devin plugin: full schema/tool parity with plugins/copilot

2026-06-25 21:56:42 +02:00

test_json_utils.py

fix(json_utils): reject prose pseudo-JSON in single quotes/backticks (#82)

2026-06-23 20:31:39 +08:00

test_materialize_searchqa.py

…

test_mcp_schema.py

feat: sync all 4 runtime plugins with full engine surface + fix #52 #58 #62

2026-06-20 11:31:09 +00:00

test_plugin_sync.py

feat: sync all 4 runtime plugins with full engine surface + fix #52 #58 #62

2026-06-20 11:31:09 +00:00

test_qwen_backend.py

…

test_scoring.py

…

test_searchqa_rollout_failfast.py

…

test_skill_aware_reflection.py

…

test_sleep_engine.py

fix(skillopt-sleep): surface codex auth/model/version failures instead of silently scoring 0 (#92)

2026-07-01 03:20:08 +08:00

test_types.py

…

test_webui_env_preflight.py

…