docs: register thinking contracts and record baseline checks
This commit is contained in:
@@ -1,3 +1,10 @@
|
||||
---
|
||||
type: design
|
||||
node_id: design:2026-09-09-134-thinking-contracts-design
|
||||
title: "1.3.4 推理意图与测试证据设计"
|
||||
date: 2026-09-09
|
||||
---
|
||||
|
||||
# 1.3.4：推理意图的可满足性与测试证据契约
|
||||
|
||||
> 日期：2026-09-09。状态：**已通过独立审查并获人类正式批准；进入实施计划阶段，编码须先完成计划审查**。
|
||||
|
||||
@@ -210,6 +210,16 @@
|
||||
"id": "plan:reasoning-effort",
|
||||
"label": "实现计划: 推理档位一等化",
|
||||
"type": "plan"
|
||||
},
|
||||
{
|
||||
"id": "design:2026-09-09-134-thinking-contracts-design",
|
||||
"label": "1.3.4 推理意图与测试证据设计",
|
||||
"type": "design"
|
||||
},
|
||||
{
|
||||
"id": "plan:2026-09-09-134-thinking-contracts",
|
||||
"label": "1.3.4 推理契约实施计划",
|
||||
"type": "plan"
|
||||
}
|
||||
],
|
||||
"links": [
|
||||
@@ -429,6 +439,13 @@
|
||||
"relation": "implements",
|
||||
"evidence": "10 个任务逐条覆盖设计 §3-§8;T10 兑现人类「能力表统一经 new-api 实测」的决定",
|
||||
"added": "2026-09-05T04:07:17.723586+00:00"
|
||||
},
|
||||
{
|
||||
"source": "plan:2026-09-09-134-thinking-contracts",
|
||||
"target": "design:2026-09-09-134-thinking-contracts-design",
|
||||
"relation": "implements",
|
||||
"evidence": "已批准设计；T0基线660 passed、make check通过",
|
||||
"added": "2026-09-09T04:48:57.560089+00:00"
|
||||
}
|
||||
]
|
||||
}
|
||||
@@ -1,8 +1,9 @@
|
||||
# Research Wiki 索引
|
||||
|
||||
> 自动生成，更新时间：2026-09-05 04:07 UTC
|
||||
> 自动生成，更新时间：2026-09-09 04:48 UTC
|
||||
|
||||
## design (41)
|
||||
## design (42)
|
||||
- [1.3.4 推理意图与测试证据设计](designs/2026-09-09-134-thinking-contracts-design.md) `design:2026-09-09-134-thinking-contracts-design`
|
||||
- [2026-07-20-m1-core-design](designs/2026-07-20-m1-core-design.md) `design:2026-07-20-m1-core-design`
|
||||
- [2026-07-20-m2-distributed-design](designs/2026-07-20-m2-distributed-design.md) `design:2026-07-20-m2-distributed-design`
|
||||
- [2026-07-21-m25-resilience-design](designs/2026-07-21-m25-resilience-design.md) `design:2026-07-21-m25-resilience-design`
|
||||
@@ -61,7 +62,8 @@
|
||||
- [P7 OCR soak 验收: 99.73% 与 13 不变量全 PASS](findings/p7-ocr-soak.md) `finding:p7-ocr-soak`
|
||||
- [推理开关与 reasoning_tokens: 供应商实测与业界做法](findings/2026-08-02-thinking-switch-and-reasoning-tokens.md) `finding:2026-08-02-thinking-switch-and-reasoning-tokens`
|
||||
|
||||
## plan (36)
|
||||
## plan (37)
|
||||
- [1.3.4 推理契约实施计划](plans/2026-09-09-134-thinking-contracts.md) `plan:2026-09-09-134-thinking-contracts`
|
||||
- [2026-07-20-m1-core-plan](plans/2026-07-20-m1-core-plan.md) `plan:2026-07-20-m1-core-plan`
|
||||
- [2026-07-20-m2-distributed-plan](plans/2026-07-20-m2-distributed-plan.md) `plan:2026-07-20-m2-distributed-plan`
|
||||
- [2026-07-21-m25-resilience-plan](plans/2026-07-21-m25-resilience-plan.md) `plan:2026-07-21-m25-resilience-plan`
|
||||
|
||||
@@ -150,3 +150,5 @@
|
||||
- [2026-09-05 04:07 UTC] 新增 plan: 实现计划: 推理档位一等化 (plan:reasoning-effort)
|
||||
- [2026-09-05 04:07 UTC] 新增边: plan:reasoning-effort --implements--> design:reasoning-effort
|
||||
- [2026-09-05 04:07 UTC] 重建索引: 95 篇页面
|
||||
- [2026-09-09 04:48 UTC] 新增边: plan:2026-09-09-134-thinking-contracts --implements--> design:2026-09-09-134-thinking-contracts-design
|
||||
- [2026-09-09 04:48 UTC] 重建索引: 97 篇页面
|
||||
|
||||
@@ -1,3 +1,10 @@
|
||||
---
|
||||
type: plan
|
||||
node_id: plan:2026-09-09-134-thinking-contracts
|
||||
title: "1.3.4 推理契约实施计划"
|
||||
date: 2026-09-09
|
||||
---
|
||||
|
||||
# 1.3.4 推理契约与测试证据实施计划
|
||||
|
||||
> 日期：2026-09-09。状态：**自审及 Codex 独立计划审查通过（复审 run ea38c3a7-12ef-4bf0-bb04-257ce37eb96f），进入执行；尚未编码／执行验收**。
|
||||
@@ -161,11 +168,11 @@ T5 按 `(session_id, parent_call_id) → AttemptEvidence.call_id → HttpEvidenc
|
||||
|
||||
### T0：基线、计划审查与文档回滚点
|
||||
|
||||
- [ ] 修改设计批准状态，新增本计划；父会话自审后前台 Codex 独立审，具体问题修正后方可执行 T1。计划无需再走人类门，不能把“已生成”当“已审”。
|
||||
- [ ] 记录 `git status --short --branch`、`git log --oneline origin/main..HEAD`、实际 HEAD；确认源代码零差异，保存未跟踪文件清单，禁止暂存 `.pi/`。
|
||||
- [ ] 执行基线：`make check`；`conda run -n PolyGateway pytest tests/unit/test_thinking.py tests/unit/test_providers.py tests/unit/test_client.py tests/unit/test_config.py tests/unit/test_openai_compat.py tests/unit/test_cache.py tests/unit/test_embedding.py tests/unit/test_ocr_client.py tests/unit/test_telemetry.py -q`。记录实际失败，不能先改期待绕过；本步骤预期现有非 slow 测试通过。
|
||||
- [ ] 以 `research-wiki` 工具登记 design／plan 节点及 implements 边，现有同路径文档不可被 add_entity 模板覆盖；先读工具已有文件处理行为，再登记、重建索引、检查生成 diff。只在本计划 writer 移交后由父会话执行这些额外文件写入。
|
||||
- [ ] 调用 commit skill，提交点 `docs: record approved thinking contracts and implementation plan`，形成生产修改前回滚点。
|
||||
- [x] 修改设计批准状态，新增本计划；父会话自审后前台 Codex 独立审，具体问题修正后方可执行 T1。计划无需再走人类门，不能把“已生成”当“已审”。
|
||||
- [x] 记录 `git status --short --branch`、`git log --oneline origin/main..HEAD`、实际 HEAD；确认源代码零差异，保存未跟踪文件清单，禁止暂存 `.pi/`。
|
||||
- [x] 执行基线：`make check`；`conda run -n PolyGateway pytest tests/unit/test_thinking.py tests/unit/test_providers.py tests/unit/test_client.py tests/unit/test_config.py tests/unit/test_openai_compat.py tests/unit/test_cache.py tests/unit/test_embedding.py tests/unit/test_ocr_client.py tests/unit/test_telemetry.py -q`。记录实际失败，不能先改期待绕过；本步骤预期现有非 slow 测试通过。
|
||||
- [x] 以 `research-wiki` 工具登记 design／plan 节点及 implements 边，现有同路径文档不可被 add_entity 模板覆盖；先读工具已有文件处理行为，再登记、重建索引、检查生成 diff。只在本计划 writer 移交后由父会话执行这些额外文件写入。
|
||||
- [x] 调用 commit skill，提交点 `docs: record approved thinking contracts and implementation plan`，形成生产修改前回滚点。
|
||||
|
||||
### T1：AUTO 成员语义与默认 MiniMax wire
|
||||
|
||||
|
||||
Reference in New Issue
Block a user