iomgaa 40b04f886e fix(question_gen): sample_anchor Codex 审查修正 ...

- C1/C2: Temporal Reasoning ≥3 L2 + Object Reasoning ≥2 L2 下限检查
- I1: L2 题型子帧不足时 ValueError
- I2: L3 过滤无 frame_path 的节点
- I3/I4: distractor_texts 扩展到整棵树范围
- I5-I7: 测试补强 Information Synopsis/Temporal Reasoning/Object Reasoning
- M1: Spatial Reasoning 测试断言 spatial_layout

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

2026-07-09 05:24:30 -04:00

..

__init__.py

build: add README, pyproject.toml, package skeletons and smoke test

2026-07-06 11:37:59 -04:00

test_agent_loop.py

feat(core/agent): 实现 AgentLoop 推理循环引擎

2026-07-06 22:43:22 -04:00

test_agent_protocols.py

feat(core/agent): 添加 ToolDispatcher Protocol 和 AgentLoopSpec hookspec

2026-07-06 22:36:31 -04:00

test_agent_types.py

feat(core/agent): 添加 Step 和 LoopResult 数据类

2026-07-06 22:32:59 -04:00

test_breaker.py

feat(adapters): 实现 CircuitBreaker 内存级熔断器

2026-07-06 22:39:57 -04:00

test_core_protocols.py

feat(core): 添加共享 Protocol 端口

2026-07-06 22:36:23 -04:00

test_core_types.py

feat(core): 追加 GeneratedQuestion frozen dataclass

2026-07-07 04:36:56 -04:00

test_diagnose.py

feat(evolution): diagnose.py aggregation + case packs + run_diagnosis (#8)

2026-07-07 10:04:48 -04:00

test_embedding_adapter.py

feat(adapters): EmbeddingProvider Protocol + local/remote 双后端实现

2026-07-07 01:45:09 -04:00

test_evolution_types.py

feat(core/evolution): protocols.py + types.py 基础层 — 18 个 dataclass + 3 个 Protocol

2026-07-07 09:34:04 -04:00

test_evolve.py

feat(evolution): evolve.py per-target evolution — skill/system/tool (#9)

2026-07-07 10:26:58 -04:00

test_gate.py

feat(evolution): gate.py — CE-Gate e-process 纯函数 (#5 算法保真)

2026-07-07 09:38:21 -04:00

test_governed_llm.py

fix(llm): call_id 移入重试循环，每 attempt 独立

2026-07-09 00:12:38 -04:00

test_harness_batching.py

feat(harness): batching.py — FFD + round-robin mini-batch (#10 算法保真)

2026-07-07 12:47:33 -04:00

test_harness_checkpoint.py

feat(harness): checkpoint.py — TrainState 序列化 + 原子写 + 指纹校验

2026-07-07 13:13:37 -04:00

test_harness_config.py

fix(config): _VIDEO_MME_TASK_TYPE_COUNT 11→12，Video-MME 实际有 12 种题型

2026-07-09 05:08:05 -04:00

test_harness_gate_ladder.py

feat(harness): gate_ladder.py — 信息阶梯 + BaselineCache (#6 算法保真)

2026-07-07 12:55:12 -04:00

test_harness_inference.py

feat(harness): inference.py — async run_inference + DI

2026-07-07 13:04:26 -04:00

test_harness_log.py

feat(harness): HarnessLog SQLite wrapper + RunLogImpl readonly port

2026-07-07 12:29:29 -04:00

test_harness_momentum.py

feat(harness): momentum.py — async 慢更新动量生成

2026-07-07 13:01:04 -04:00

test_harness_observation.py

feat(harness): observation.py — 五张观测表 + step/epoch 报告

2026-07-07 12:55:12 -04:00

test_harness_pools.py

feat(harness): pools.py — 三池切分（test→validation→diagnosis）

2026-07-07 12:48:54 -04:00

test_harness_runner.py

feat(harness): runner.py — train loop orchestrator (#13 algorithm fidelity)

2026-07-07 13:43:20 -04:00

test_harness_store.py

feat(harness): store.py — Store 版本操作 + Seed 管理

2026-07-07 12:36:36 -04:00

test_harness_validate.py

feat(harness): validate.py — async 块序贯验证编排 + Probation 统一定义

2026-07-07 13:20:43 -04:00

test_harness_workspace.py

feat(harness): workspace.py — Workspace lifecycle + VersionedSkillStore/PromptStore

2026-07-07 12:41:59 -04:00

test_ocr_adapter.py

feat(search): __init__.py 公开 API + 修复 OCR 测试 asyncio 兼容性

2026-07-07 06:13:25 -04:00

test_patch.py

feat(evolution): patch.py — 补丁引擎移植（算法 #9），51 测试全通过

2026-07-07 09:44:05 -04:00

test_question_gen_api.py

feat(question_gen): QuestionGenerator Protocol + 模块公开 API

2026-07-07 04:48:21 -04:00

test_question_loader.py

feat(question_gen): stratified_sample — 分层采样 + 题型保底

2026-07-07 04:45:19 -04:00

test_redis_cache.py

feat(adapters): 实现 RedisResponseCache

2026-07-06 22:39:42 -04:00

test_repair_detector.py

fix(detector): 移除 ongoing_actions/visible_entities 空字段检测 + 新增 repair sh

2026-07-09 00:54:27 -04:00

test_repair_progress.py

feat(repair): 断点续跑 progress 文件管理

2026-07-09 00:23:04 -04:00

test_repair_supplement.py

feat(tree/repair): Q&A 反向补全 — 从 TRM4 supplement 迁移

2026-07-07 02:37:50 -04:00

test_search_prompt.py

feat(search): PromptManager — 搜索 Agent 提示词加载与组装

2026-07-07 06:11:13 -04:00

test_search_skills.py

feat(search): app/search/skills.py — 技能注册表与 frontmatter 解析

2026-07-07 05:46:25 -04:00

test_search_summarizer.py

style: format vision.py

2026-07-07 05:53:33 -04:00

test_search_tools.py

feat(search): 实现 SearchToolDispatcher 工具调度器 (Task 7)

2026-07-07 06:07:27 -04:00

test_search_vision.py

feat(search): app/search/vision.py — 两轮 VLM 帧观察模块

2026-07-07 05:52:54 -04:00

test_smoke.py

build: add README, pyproject.toml, package skeletons and smoke test

2026-07-06 11:37:59 -04:00

test_streaming.py

feat(adapters): 实现三层流式看门狗

2026-07-06 22:43:01 -04:00

test_subtitle.py

feat(tree): 字幕模块 — SRT 解析 + 完整性检查 + Voronoi 分配

2026-07-07 01:54:09 -04:00

test_synthesizer.py

fix(question_gen): sample_anchor Codex 审查修正

2026-07-09 05:24:30 -04:00

test_telemetry.py

fix(telemetry): INSERT OR IGNORE + WAL + try/except 三层防御加固

2026-07-09 00:11:54 -04:00

test_tree_environment.py

feat(tree): add get_node_text + get_children_info to TreeEnvironment

2026-07-07 05:45:48 -04:00

test_tree_index.py

feat(tree): TreeIndex 数据结构 — Card 体系 + 节点 + 序列化

2026-07-07 01:35:10 -04:00

test_validate.py

feat(evolution): validate.py — pure block validation decision functions (#7)

2026-07-07 09:47:45 -04:00

test_verify.py

feat(tree): 质量校验 — 交叉验证 entities/visible_text

2026-07-07 01:57:32 -04:00

test_video_builder.py

feat(tree): VideoTreeBuilder 保真 #1 #2 #3 + 复杂度重构

2026-07-07 02:17:50 -04:00

test_vlm_adapter.py

feat(adapters): GovernedVLMClient — VLMProvider 最小可用实现

2026-07-07 01:49:55 -04:00