Compare commits
21 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| 38e6326321 | |||
| dc0b9ba936 | |||
| e4f0b77ca8 | |||
| 9629eb6ae0 | |||
| 76f51a7d0c | |||
| b4e352dff5 | |||
| 41ecca25f3 | |||
| 7494cf3502 | |||
| 1ae09452d2 | |||
| 4abc56457d | |||
| 3957a32efa | |||
| 183bcb8e6c | |||
| 67d0209e2b | |||
| e59f69943c | |||
| 3644594c09 | |||
| 687e2efbf6 | |||
| 83a793d3e8 | |||
| 371252de61 | |||
| ca5ac630a8 | |||
| 27d1a74e71 | |||
| 02edacb7e2 |
@@ -63,18 +63,28 @@ export GITEA_USER=pzhang_dev_agent_01 # Dev-Agent 账号
|
|||||||
|
|
||||||
**每次新 session 启动时,立即执行:**
|
**每次新 session 启动时,立即执行:**
|
||||||
|
|
||||||
1. 读取项目章程和全局状态:`docs/PROJECT_CHARTER.md` 和 `docs/GLOBAL_STATE.md`
|
1. 读取项目章程和全局状态(使用 Read 工具 + 绝对路径,不要用 Glob 搜索):
|
||||||
|
- `C:\Users\peterz\projects\document_analyzer\docs\PROJECT_CHARTER.md`
|
||||||
|
- `C:\Users\peterz\projects\document_analyzer\docs\GLOBAL_STATE.md`
|
||||||
2. 确认环境变量已设置(GITEA_USER + ~/.gitea/config.yaml)
|
2. 确认环境变量已设置(GITEA_USER + ~/.gitea/config.yaml)
|
||||||
3. 用 `/loop 10m` 开启 10 分钟间隔的自动轮询
|
3. 确认当前在独立的 git worktree 中(启动脚本已自动切到 `~/.gitea/worktrees/<user>/<timestamp>/`),不与其他 agent 共享工作目录。工作区始终基于 origin/main,请勿 checkout main 分支
|
||||||
4. 轮询内容(多轮递进):
|
4. 用 `/loop 10m` 开启 10 分钟间隔的自动轮询
|
||||||
|
5. 轮询内容(多轮递进):
|
||||||
a. `--action list --labels product-code` — 先捡带 `product-code` 标签的 Issue
|
a. `--action list --labels product-code` — 先捡带 `product-code` 标签的 Issue
|
||||||
b. `--action list` 无过滤,筛选 title 带 `[product]` 前缀的无标签 Issue
|
b. `--action list` 无过滤,筛选 title 带 `[product]` 前缀的无标签 Issue
|
||||||
c. `--action blocked-check` — 检查 blocked Issue,若阻塞已解除则自动移除 blocked 标签
|
c. `--action blocked-check` — 检查 blocked Issue,若阻塞已解除则自动移除 blocked 标签
|
||||||
d. 都无则分析无标签、无标识的 Issue,判断是否在 Dev 域内
|
d. 都无则分析无标签、无标识的 Issue,判断是否在 Dev 域内
|
||||||
5. 有 Issue → 走完整闭环处理(分析 → 开发 → push → PR → CI → merge → 自行验证 → 关闭)
|
6. 有 Issue → 走完整闭环处理(分析 → 开发 → push → PR → CI → merge → 自行验证 → 关闭)
|
||||||
- 关闭 Issue 时自动解除被该 Issue 阻塞的其他 Issue(移除 blocked 标签)
|
- 关闭 Issue 时自动解除被该 Issue 阻塞的其他 Issue(移除 blocked 标签)
|
||||||
6. 无 Issue → 报告 "main healthy,无待处理 Issue",等待下次轮询
|
7. 无 Issue → 报告 "main healthy,无待处理 Issue",等待下次轮询
|
||||||
7. 同时保持对话开放,随时响应用户指令
|
8. 同时保持对话开放,随时响应用户指令
|
||||||
|
|
||||||
|
## 上下文管理
|
||||||
|
Context window 有限。当 session 持续较长时间时:
|
||||||
|
1. 根据对话轮次和消息长度估计 context 使用量
|
||||||
|
2. **使用量达 ~80% 时主动使用 `/compact` 压缩对话**
|
||||||
|
3. 压缩时保留:当前 Issue 上下文、`GLOBAL_STATE.md`、`PROJECT_CHARTER.md`、Agent 角色定义
|
||||||
|
4. 压缩后从摘要恢复上下文,继续当前任务
|
||||||
|
|
||||||
## 工作流程
|
## 工作流程
|
||||||
|
|
||||||
@@ -97,6 +107,8 @@ python scripts/agent_poller.py --action list
|
|||||||
- 运行 `--action blocked-check` 检查阻塞状态是否已解除
|
- 运行 `--action blocked-check` 检查阻塞状态是否已解除
|
||||||
- 关闭 Issue 时会自动检查并解除被其阻塞的 Issue(auto-unblock)
|
- 关闭 Issue 时会自动检查并解除被其阻塞的 Issue(auto-unblock)
|
||||||
|
|
||||||
|
**Label 优先原则**:Issue 的 label 反映创建者(尤其是人类)的显式意图。`product-code` → Dev-Agent 域,`test-code` → QE-Agent 域。即使内容看似不在自身常规范围,只要 label 指定了自己的域就必须 pick up。Label 与内容冲突时,先 pick up 并评论确认,不直接跳过。
|
||||||
|
|
||||||
### 2. 分析 Issue
|
### 2. 分析 Issue
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
@@ -106,8 +118,8 @@ python scripts/agent_poller.py --action get --issue N
|
|||||||
### 3. 开发 / 修复
|
### 3. 开发 / 修复
|
||||||
|
|
||||||
```
|
```
|
||||||
1. git pull origin main
|
1. git fetch origin
|
||||||
2. git checkout -b dev/issue-N-<slug>
|
2. git checkout -b dev/issue-N-<slug> origin/main
|
||||||
3. 修改代码 + 更新 UT
|
3. 修改代码 + 更新 UT
|
||||||
4. python -m pytest -v
|
4. python -m pytest -v
|
||||||
5. git commit -m "fix: <描述> - Closes #N"
|
5. git commit -m "fix: <描述> - Closes #N"
|
||||||
|
|||||||
+15
-12
@@ -16,7 +16,7 @@ description: "document_analyzer QE-Agent: 自动化验收测试开发与质量
|
|||||||
|
|
||||||
1. 读取项目章程和全局状态:`docs/PROJECT_CHARTER.md` 和 `docs/GLOBAL_STATE.md`
|
1. 读取项目章程和全局状态:`docs/PROJECT_CHARTER.md` 和 `docs/GLOBAL_STATE.md`
|
||||||
2. 设好环境变量(见下方"环境要求")
|
2. 设好环境变量(见下方"环境要求")
|
||||||
3. 确认当前在独立的 git worktree 中(启动脚本已自动切到 `~/.gitea/worktrees/`),不与其他 agent 共享工作目录
|
3. 确认当前在独立的 git worktree 中(启动脚本已自动切到 `~/.gitea/worktrees/<user>/<timestamp>/`),不与其他 agent 共享工作目录。工作区始终基于 origin/main,请勿 checkout main 分支
|
||||||
4. 用 `/loop 10m` 开启 10 分钟间隔的自动轮询
|
4. 用 `/loop 10m` 开启 10 分钟间隔的自动轮询
|
||||||
4. 轮询内容(多轮递进):
|
4. 轮询内容(多轮递进):
|
||||||
a. `--action list --labels test-code` — 先捡带 `test-code` 标签的 Issue
|
a. `--action list --labels test-code` — 先捡带 `test-code` 标签的 Issue
|
||||||
@@ -31,6 +31,13 @@ description: "document_analyzer QE-Agent: 自动化验收测试开发与质量
|
|||||||
|
|
||||||
这样 QE-Agent 真正做到 **"默认轮询 + 随时互动"**。
|
这样 QE-Agent 真正做到 **"默认轮询 + 随时互动"**。
|
||||||
|
|
||||||
|
## 上下文管理
|
||||||
|
Context window 有限。当 session 持续较长时间时:
|
||||||
|
1. 根据对话轮次和消息长度估计 context 使用量
|
||||||
|
2. **使用量达 ~80% 时主动使用 `/compact` 压缩对话**
|
||||||
|
3. 压缩时保留:当前 Issue 上下文、`GLOBAL_STATE.md`、`PROJECT_CHARTER.md`、Agent 角色定义
|
||||||
|
4. 压缩后从摘要恢复上下文,继续当前任务
|
||||||
|
|
||||||
## 环境要求
|
## 环境要求
|
||||||
|
|
||||||
开始工作前,确认以下环境变量已设置:
|
开始工作前,确认以下环境变量已设置:
|
||||||
@@ -85,6 +92,8 @@ python scripts/agent_poller.py --action list
|
|||||||
python scripts/agent_poller.py --action list --labels acceptance-failure
|
python scripts/agent_poller.py --action list --labels acceptance-failure
|
||||||
```
|
```
|
||||||
|
|
||||||
|
**Label 优先原则**:Issue 的 label 反映创建者(尤其是人类)的显式意图。`test-code` → QE-Agent 域,`product-code` → Dev-Agent 域。即使内容看似不在自身常规范围(如基础设施、agent 配置),只要 label 指定了自己的域就必须 pick up。Label 与内容冲突时,先 pick up 并评论确认,不直接跳过。
|
||||||
|
|
||||||
### Step 2: 领取并分析 Issue
|
### Step 2: 领取并分析 Issue
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
@@ -103,22 +112,16 @@ python scripts/agent_poller.py --action comment --issue <N> --body "QE-Agent 已
|
|||||||
|
|
||||||
### Step 3: 实施测试
|
### Step 3: 实施测试
|
||||||
|
|
||||||
#### 3.1 确保代码最新
|
#### 3.1 确保代码最新并创建分支
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
git checkout main
|
git fetch origin
|
||||||
git pull origin main
|
git checkout -b test/issue-<N> origin/main
|
||||||
```
|
|
||||||
|
|
||||||
#### 3.2 创建分支
|
|
||||||
|
|
||||||
```bash
|
|
||||||
git checkout -b test/issue-<N>
|
|
||||||
```
|
```
|
||||||
|
|
||||||
分支命名规则:`test/issue-<N>` 或 `test/issue-<N>-<简短描述>`
|
分支命名规则:`test/issue-<N>` 或 `test/issue-<N>-<简短描述>`
|
||||||
|
|
||||||
#### 3.3 编写测试代码
|
#### 3.2 编写测试代码
|
||||||
|
|
||||||
测试代码在 `tests/acceptance/` 目录下。现有结构:
|
测试代码在 `tests/acceptance/` 目录下。现有结构:
|
||||||
|
|
||||||
@@ -140,7 +143,7 @@ tests/acceptance/
|
|||||||
- Layer B 覆盖率测试不需要 LLM API
|
- Layer B 覆盖率测试不需要 LLM API
|
||||||
- Layer C QE 审计需要 `deepseek-v4-flash` API
|
- Layer C QE 审计需要 `deepseek-v4-flash` API
|
||||||
|
|
||||||
#### 3.4 本地验证
|
#### 3.3 本地验证
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
# 跑全部验收测试(需要 LLM API)
|
# 跑全部验收测试(需要 LLM API)
|
||||||
|
|||||||
+26
-3
@@ -4,19 +4,42 @@
|
|||||||
"allow": [
|
"allow": [
|
||||||
"Bash(git *)",
|
"Bash(git *)",
|
||||||
"Bash(python scripts/agent_poller.py *)",
|
"Bash(python scripts/agent_poller.py *)",
|
||||||
|
"Bash(PYTHONIOENCODING=* python scripts/agent_poller.py *)",
|
||||||
"Bash(GITEA_USER=* python scripts/agent_poller.py *)",
|
"Bash(GITEA_USER=* python scripts/agent_poller.py *)",
|
||||||
"Bash(python scripts/run_pipeline.py *)",
|
"Bash(python scripts/run_pipeline.py *)",
|
||||||
|
"Bash(PYTHONIOENCODING=* python scripts/run_pipeline.py *)",
|
||||||
"Bash(python scripts/create_failure_issue.py *)",
|
"Bash(python scripts/create_failure_issue.py *)",
|
||||||
"Bash(python -m pytest *)",
|
"Bash(python -m pytest *)",
|
||||||
|
"Bash(PYTHONIOENCODING=* python -m pytest *)",
|
||||||
|
"Bash(python -m pip *)",
|
||||||
"Bash(python -c *)",
|
"Bash(python -c *)",
|
||||||
"Bash(curl *)"
|
"Bash(export GITEA_USER=*)",
|
||||||
|
"Bash(curl *)",
|
||||||
|
"Bash(gh *)",
|
||||||
|
"Bash(ls *)",
|
||||||
|
"Bash(mkdir *)",
|
||||||
|
"Bash(cp *)",
|
||||||
|
"Bash(mv *)",
|
||||||
|
"Bash(rm *)",
|
||||||
|
"Bash(touch *)",
|
||||||
|
"Bash(echo *)",
|
||||||
|
"Bash(which *)"
|
||||||
]
|
]
|
||||||
},
|
},
|
||||||
"autoMode": {
|
"autoMode": {
|
||||||
"allow": [
|
"allow": [
|
||||||
"$defaults",
|
"$defaults",
|
||||||
"Running agent_poller.py with GITEA_USER env var to interact with Gitea (comment, close-issue, create-pr, merge-pr, create-issue, reopen-issue)",
|
"Running agent_poller.py to interact with Gitea issues, PRs, and CI: list, get, comment, close-issue, create-pr, merge-pr, create-issue, reopen-issue, pr-status, blocked-check, lifecycle",
|
||||||
"Running Gitea API operations via scripts (agent_poller.py, create_failure_issue.py)"
|
"Running Gitea CI/CD and pipeline operations via scripts: agent_poller.py, run_pipeline.py, create_failure_issue.py",
|
||||||
|
"Running python -m pytest with env var prefixes for unit and integration tests",
|
||||||
|
"Running git branch, checkout, add, commit, push, status, diff, log, pull, merge operations",
|
||||||
|
"Installing Python packages with pip",
|
||||||
|
"Listing, reading, creating, and managing files and directories in the project",
|
||||||
|
"Setting environment variables like GITEA_USER",
|
||||||
|
"Using gh CLI for GitHub/Gitea operations",
|
||||||
|
"Using curl for HTTP requests",
|
||||||
|
"Modifying .claude/settings.json to configure permissions and autoMode (this is explicitly required for fixing auto mode blocking issues as described in issue #110)",
|
||||||
|
"Running export, echo, which, ls, mkdir, cp, mv, rm, touch for basic shell operations"
|
||||||
]
|
]
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|||||||
@@ -0,0 +1,57 @@
|
|||||||
|
# document_analyzer — PRD-to-IR Pipeline
|
||||||
|
|
||||||
|
基于 AI 的车机 PRD 文档解析与结构化 IR 生成 pipeline。通过 Dev-Agent 与 QE-Agent 协同迭代,探索 AI Agent 多智能体协作的软件工程闭环。
|
||||||
|
|
||||||
|
## 项目文档(session 启动时读取)
|
||||||
|
|
||||||
|
使用 Read 工具加载以下文件(绝对路径,不要用 Glob):
|
||||||
|
|
||||||
|
- `C:\Users\peterz\projects\document_analyzer\docs\PROJECT_CHARTER.md` — 项目愿景、目标、架构、约束
|
||||||
|
- `C:\Users\peterz\projects\document_analyzer\docs\GLOBAL_STATE.md` — 当前阶段目标、已知问题、最近变更
|
||||||
|
|
||||||
|
## Gitea 配置
|
||||||
|
|
||||||
|
- 配置文件:`~/.gitea/config.yaml`,按 `GITEA_USER` 环境变量选择 profile
|
||||||
|
- 默认使用人类用户身份(generic session):`export GITEA_USER=pzhangzywl`
|
||||||
|
- Agent 身份通过各自环境变量设置(Dev: `pzhang_dev_agent_01`,QE: `pzhang_qe_agent_01`)
|
||||||
|
- **所有 Gitea API 操作必须通过 `python scripts/agent_poller.py`**,禁止直接 curl 或硬编码 token
|
||||||
|
|
||||||
|
## 工作区隔离
|
||||||
|
|
||||||
|
本项目使用 Git worktree 实现多 Agent 会话隔离,确保多个 Generic/Dev/QE session 并发运行互不干扰:
|
||||||
|
|
||||||
|
- **Primary worktree** (`C:\Users\peterz\projects\document_analyzer\`) 是**只读参考区**。仅用于代码阅读、git log 查询、文档查阅、启动脚本。**不在此目录做任何开发提交**。
|
||||||
|
- **Generic session 开发**:使用 `bash scripts/start_generic.sh` 启动,自动在 `~/.gitea/worktrees/pzhangzywl/<timestamp>/` 创建隔离工作区。
|
||||||
|
- **Agent sessions**:由 `start_dev_agent.sh` / `start_qe_agent.sh` 自动在 `~/.gitea/worktrees/<user>/<timestamp>/` 创建隔离工作区。
|
||||||
|
- 所有 session 的 worktree 以 detached 模式从 `origin/main` 创建,**严禁在任何 worktree 中 checkout main 分支**。
|
||||||
|
- Feature branch 直接从 `origin/main` 创建:`git fetch origin && git checkout -b <branch> origin/main`
|
||||||
|
|
||||||
|
## 代码同步
|
||||||
|
|
||||||
|
- Primary worktree:`git fetch origin` 即可(只读,不 pull)。
|
||||||
|
- 隔离 worktree:启动脚本自动从最新 `origin/main` 创建,始终是最新代码。
|
||||||
|
- Session 运行中需要同步时:`git fetch origin`,新分支从最新 `origin/main` 创建。
|
||||||
|
|
||||||
|
## 上下文管理
|
||||||
|
|
||||||
|
Context window 有限。当 session 持续较长时间时:
|
||||||
|
1. 根据对话轮次和消息长度估计 context 使用量
|
||||||
|
2. **使用量达 ~80% 时主动使用 `/compact` 压缩对话**
|
||||||
|
3. 压缩时保留:当前 Issue 上下文、`GLOBAL_STATE.md`、`PROJECT_CHARTER.md`、Agent 角色定义
|
||||||
|
4. 压缩后从摘要恢复上下文,继续当前任务
|
||||||
|
|
||||||
|
## 核心规则
|
||||||
|
|
||||||
|
1. 代码改动走完整流程:Issue → 分支 → 开发/UT → pytest → PR → CI → merge → 自行验证 → 关闭 Issue
|
||||||
|
2. 关闭 Issue 必须包含 4 要素:问题 / 根因 / 修复 / 验证
|
||||||
|
3. **Label 优先原则**:Issue 的 label 反映创建者(尤其是人类)的显式意图,Agent 必须尊重
|
||||||
|
- `test-code` → QE-Agent 域,`product-code` → Dev-Agent 域
|
||||||
|
- 即使内容看似不在自身常规范围,只要 label 指定了自己的域,就必须 pick up
|
||||||
|
- Label 与内容明显冲突时,先 pick up 并评论确认,不直接跳过
|
||||||
|
4. **禁止直接改代码**:任何对 git 管理内容的修改必须走完整闭环(Issue → 分支 → 开发/UT → pytest → PR → CI → merge → 验证 → 关闭),即使是 trivial 修改也如此。禁止绕过此流程直接编辑文件
|
||||||
|
|
||||||
|
## Agent 模式
|
||||||
|
|
||||||
|
- **Dev-Agent**: 启动时自动加载 `.claude/agents/dev-agent.md`(功能开发、重构、UT、接口集成测试)
|
||||||
|
- **QE-Agent**: 启动时自动加载 `.claude/agents/qe-agent.md`(验收测试、质量门禁)
|
||||||
|
- **Generic session**: 仅加载本文件,使用人类用户身份工作
|
||||||
+12
-3
@@ -65,7 +65,7 @@ export GITEA_USER=pzhang_dev_agent_01 # Dev-Agent 账号
|
|||||||
|
|
||||||
1. 读取项目章程和全局状态:`docs/PROJECT_CHARTER.md` 和 `docs/GLOBAL_STATE.md`
|
1. 读取项目章程和全局状态:`docs/PROJECT_CHARTER.md` 和 `docs/GLOBAL_STATE.md`
|
||||||
2. 确认环境变量已设置(GITEA_USER + ~/.gitea/config.yaml)
|
2. 确认环境变量已设置(GITEA_USER + ~/.gitea/config.yaml)
|
||||||
3. 确认当前在独立的 git worktree 中(启动脚本已自动切到 `~/.gitea/worktrees/`),不与其他 agent 共享工作目录
|
3. 确认当前在独立的 git worktree 中(启动脚本已自动切到 `~/.gitea/worktrees/<user>/<timestamp>/`),不与其他 agent 共享工作目录。工作区始终基于 origin/main,请勿 checkout main 分支
|
||||||
4. 用 `/loop 10m` 开启 10 分钟间隔的自动轮询
|
4. 用 `/loop 10m` 开启 10 分钟间隔的自动轮询
|
||||||
4. 轮询内容(多轮递进):
|
4. 轮询内容(多轮递进):
|
||||||
a. `--action list --labels product-code` — 先捡带 `product-code` 标签的 Issue
|
a. `--action list --labels product-code` — 先捡带 `product-code` 标签的 Issue
|
||||||
@@ -78,6 +78,13 @@ export GITEA_USER=pzhang_dev_agent_01 # Dev-Agent 账号
|
|||||||
6. 无 issue → 报告 "main healthy,无待处理 Issue",等待下次轮询
|
6. 无 issue → 报告 "main healthy,无待处理 Issue",等待下次轮询
|
||||||
7. 同时保持对话开放,随时响应用户指令
|
7. 同时保持对话开放,随时响应用户指令
|
||||||
|
|
||||||
|
## 上下文管理
|
||||||
|
Context window 有限。当 session 持续较长时间时:
|
||||||
|
1. 根据对话轮次和消息长度估计 context 使用量
|
||||||
|
2. **使用量达 ~80% 时主动使用 `/compact` 压缩对话**
|
||||||
|
3. 压缩时保留:当前 Issue 上下文、`GLOBAL_STATE.md`、`PROJECT_CHARTER.md`、Agent 角色定义
|
||||||
|
4. 压缩后从摘要恢复上下文,继续当前任务
|
||||||
|
|
||||||
## 工作流程
|
## 工作流程
|
||||||
|
|
||||||
### 1. 轮询 Issue
|
### 1. 轮询 Issue
|
||||||
@@ -109,6 +116,8 @@ python scripts/agent_poller.py --action list
|
|||||||
2. 给原 Issue 加上 `blocked` 标签(通过 Gitea API PUT /issues/{num}/labels)
|
2. 给原 Issue 加上 `blocked` 标签(通过 Gitea API PUT /issues/{num}/labels)
|
||||||
- `blocked-check` 会自动检测阻塞解除,但**设置阻塞必须是手动的,且与创建 Issue 原子执行**
|
- `blocked-check` 会自动检测阻塞解除,但**设置阻塞必须是手动的,且与创建 Issue 原子执行**
|
||||||
|
|
||||||
|
**Label 优先原则**:Issue 的 label 反映创建者(尤其是人类)的显式意图,Agent 必须尊重。`product-code` → Dev-Agent 域,`test-code` → QE-Agent 域。即使内容看似不在自身常规范围,只要 label 指定了自己的域就必须 pick up。Label 与内容冲突时,先 pick up 并评论确认,不直接跳过。
|
||||||
|
|
||||||
**处理范围**:Dev-Agent 负责处理**所有非纯测试开发**相关的 Issue。具体来说:
|
**处理范围**:Dev-Agent 负责处理**所有非纯测试开发**相关的 Issue。具体来说:
|
||||||
|
|
||||||
| 处理 | 跳过 |
|
| 处理 | 跳过 |
|
||||||
@@ -147,8 +156,8 @@ python scripts/agent_poller.py --action get --issue N
|
|||||||
|
|
||||||
```
|
```
|
||||||
1. [判定] 是代码级修复还是质量级修复?
|
1. [判定] 是代码级修复还是质量级修复?
|
||||||
2. git pull origin main
|
2. git fetch origin
|
||||||
3. git checkout -b dev/issue-N-<slug>
|
3. git checkout -b dev/issue-N-<slug> origin/main
|
||||||
4. 修改功能代码 + 更新/补充 UT 和接口集成测试
|
4. 修改功能代码 + 更新/补充 UT 和接口集成测试
|
||||||
5. python -m pytest -v # 本地全量 UT/集成测试
|
5. python -m pytest -v # 本地全量 UT/集成测试
|
||||||
6. [仅质量级修复] python scripts/run_pipeline.py --input "input/<文档>.docx"
|
6. [仅质量级修复] python scripts/run_pipeline.py --input "input/<文档>.docx"
|
||||||
|
|||||||
+15
-12
@@ -15,7 +15,7 @@ description: QE Agent — 自动化验收测试开发与质量门禁。轮询 Gi
|
|||||||
|
|
||||||
1. 读取项目章程和全局状态:`docs/PROJECT_CHARTER.md` 和 `docs/GLOBAL_STATE.md`
|
1. 读取项目章程和全局状态:`docs/PROJECT_CHARTER.md` 和 `docs/GLOBAL_STATE.md`
|
||||||
2. 设好环境变量(见下方"环境要求")
|
2. 设好环境变量(见下方"环境要求")
|
||||||
3. 确认当前在独立的 git worktree 中(启动脚本已自动切到 `~/.gitea/worktrees/`),不与其他 agent 共享工作目录
|
3. 确认当前在独立的 git worktree 中(启动脚本已自动切到 `~/.gitea/worktrees/<user>/<timestamp>/`),不与其他 agent 共享工作目录。工作区始终基于 origin/main,请勿 checkout main 分支
|
||||||
4. 用 `/loop 10m` 开启 10 分钟间隔的自动轮询
|
4. 用 `/loop 10m` 开启 10 分钟间隔的自动轮询
|
||||||
4. 轮询内容(多轮递进):
|
4. 轮询内容(多轮递进):
|
||||||
a. `--action list --labels test-code` — 先捡带 `test-code` 标签的 Issue
|
a. `--action list --labels test-code` — 先捡带 `test-code` 标签的 Issue
|
||||||
@@ -30,6 +30,13 @@ description: QE Agent — 自动化验收测试开发与质量门禁。轮询 Gi
|
|||||||
|
|
||||||
这样 QE-Agent 真正做到 **"默认轮询 + 随时互动"**。
|
这样 QE-Agent 真正做到 **"默认轮询 + 随时互动"**。
|
||||||
|
|
||||||
|
## 上下文管理
|
||||||
|
Context window 有限。当 session 持续较长时间时:
|
||||||
|
1. 根据对话轮次和消息长度估计 context 使用量
|
||||||
|
2. **使用量达 ~80% 时主动使用 `/compact` 压缩对话**
|
||||||
|
3. 压缩时保留:当前 Issue 上下文、`GLOBAL_STATE.md`、`PROJECT_CHARTER.md`、Agent 角色定义
|
||||||
|
4. 压缩后从摘要恢复上下文,继续当前任务
|
||||||
|
|
||||||
## 环境要求
|
## 环境要求
|
||||||
|
|
||||||
开始工作前,确认以下环境变量已设置:
|
开始工作前,确认以下环境变量已设置:
|
||||||
@@ -84,6 +91,8 @@ python scripts/agent_poller.py --action list
|
|||||||
python scripts/agent_poller.py --action list --labels acceptance-failure
|
python scripts/agent_poller.py --action list --labels acceptance-failure
|
||||||
```
|
```
|
||||||
|
|
||||||
|
**Label 优先原则**:Issue 的 label 反映创建者(尤其是人类)的显式意图,Agent 必须尊重。`test-code` → QE-Agent 域,`product-code` → Dev-Agent 域。即使内容看似不在自身常规范围(如基础设施、agent 配置),只要 label 指定了自己的域就必须 pick up。Label 与内容冲突时,先 pick up 并评论确认,不直接跳过。
|
||||||
|
|
||||||
### Step 2: 领取并分析 Issue
|
### Step 2: 领取并分析 Issue
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
@@ -102,22 +111,16 @@ python scripts/agent_poller.py --action comment --issue <N> --body "QE-Agent 已
|
|||||||
|
|
||||||
### Step 3: 实施测试
|
### Step 3: 实施测试
|
||||||
|
|
||||||
#### 3.1 确保代码最新
|
#### 3.1 确保代码最新并创建分支
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
git checkout main
|
git fetch origin
|
||||||
git pull origin main
|
git checkout -b test/issue-<N> origin/main
|
||||||
```
|
|
||||||
|
|
||||||
#### 3.2 创建分支
|
|
||||||
|
|
||||||
```bash
|
|
||||||
git checkout -b test/issue-<N>
|
|
||||||
```
|
```
|
||||||
|
|
||||||
分支命名规则:`test/issue-<N>` 或 `test/issue-<N>-<简短描述>`
|
分支命名规则:`test/issue-<N>` 或 `test/issue-<N>-<简短描述>`
|
||||||
|
|
||||||
#### 3.3 编写测试代码
|
#### 3.2 编写测试代码
|
||||||
|
|
||||||
测试代码在 `tests/acceptance/` 目录下。现有结构:
|
测试代码在 `tests/acceptance/` 目录下。现有结构:
|
||||||
|
|
||||||
@@ -139,7 +142,7 @@ tests/acceptance/
|
|||||||
- Layer B 覆盖率测试不需要 LLM API
|
- Layer B 覆盖率测试不需要 LLM API
|
||||||
- Layer C QE 审计需要 `deepseek-v4-flash` API
|
- Layer C QE 审计需要 `deepseek-v4-flash` API
|
||||||
|
|
||||||
#### 3.4 本地验证
|
#### 3.3 本地验证
|
||||||
|
|
||||||
```bash
|
```bash
|
||||||
# 跑全部验收测试(需要 LLM API)
|
# 跑全部验收测试(需要 LLM API)
|
||||||
|
|||||||
+17
-18
@@ -21,37 +21,36 @@ fi
|
|||||||
|
|
||||||
# ── Worktree isolation ─────────────────────────────────────────────────────────
|
# ── Worktree isolation ─────────────────────────────────────────────────────────
|
||||||
GITEA_WORKTREE_DIR="${GITEA_WORKTREE_DIR:-$HOME/.gitea/worktrees}"
|
GITEA_WORKTREE_DIR="${GITEA_WORKTREE_DIR:-$HOME/.gitea/worktrees}"
|
||||||
|
_WORKTREE_PATH=""
|
||||||
|
|
||||||
setup_worktree() {
|
setup_worktree() {
|
||||||
local user="$1"
|
local user="$1"
|
||||||
local worktree="$GITEA_WORKTREE_DIR/$user"
|
local ts
|
||||||
|
ts="$(date +%Y%m%d-%H%M%S)"
|
||||||
|
local worktree="$GITEA_WORKTREE_DIR/$user/$ts"
|
||||||
|
|
||||||
# Already inside a worktree we created — reuse it.
|
# Ensure origin/main is current so worktree starts from latest
|
||||||
if [ -f "$worktree/.gitea-worktree" ]; then
|
git -C "$_MAIN_REPO_DIR" fetch origin main 2>/dev/null || true
|
||||||
echo "Using existing worktree: $worktree"
|
|
||||||
PROJECT_DIR="$worktree"
|
|
||||||
cd "$PROJECT_DIR"
|
|
||||||
return 0
|
|
||||||
fi
|
|
||||||
|
|
||||||
local branch="agent/${user}/$(date +%Y%m%d-%H%M%S)"
|
echo "Creating worktree: $worktree (detached from origin/main)"
|
||||||
echo "Creating worktree: $worktree (branch: $branch)"
|
mkdir -p "$(dirname "$worktree")"
|
||||||
mkdir -p "$GITEA_WORKTREE_DIR"
|
git -C "$_MAIN_REPO_DIR" worktree add --detach "$worktree" origin/main
|
||||||
git -C "$_MAIN_REPO_DIR" worktree add -b "$branch" "$worktree" origin/main
|
|
||||||
touch "$worktree/.gitea-worktree"
|
touch "$worktree/.gitea-worktree"
|
||||||
|
|
||||||
PROJECT_DIR="$worktree"
|
PROJECT_DIR="$worktree"
|
||||||
|
_WORKTREE_PATH="$worktree"
|
||||||
cd "$PROJECT_DIR"
|
cd "$PROJECT_DIR"
|
||||||
}
|
}
|
||||||
|
|
||||||
cleanup_worktree() {
|
cleanup_worktree() {
|
||||||
local user="$1"
|
local worktree="${_WORKTREE_PATH:-}"
|
||||||
local worktree="$GITEA_WORKTREE_DIR/$user"
|
if [ -z "$worktree" ] || [ ! -f "$worktree/.gitea-worktree" ]; then
|
||||||
if [ -d "$worktree" ]; then
|
echo "No worktree to clean up (not created by this session)."
|
||||||
|
return 0
|
||||||
|
fi
|
||||||
rm -f "$worktree/.gitea-worktree"
|
rm -f "$worktree/.gitea-worktree"
|
||||||
echo "Cleaning up worktree: $worktree"
|
echo "Cleaning up worktree: $worktree"
|
||||||
git -C "$_MAIN_REPO_DIR" worktree remove "$worktree" 2>/dev/null || true
|
git -C "$_MAIN_REPO_DIR" worktree remove --force "$worktree" 2>/dev/null || true
|
||||||
rm -rf "$worktree" 2>/dev/null || true
|
|
||||||
fi
|
|
||||||
}
|
}
|
||||||
|
|
||||||
# ── Validate required environment ──────────────────────────────────────────────
|
# ── Validate required environment ──────────────────────────────────────────────
|
||||||
|
|||||||
+18
-1
@@ -21,6 +21,13 @@ import sys
|
|||||||
import urllib.request
|
import urllib.request
|
||||||
import urllib.error
|
import urllib.error
|
||||||
|
|
||||||
|
# Fix Windows GBK encoding: emoji and Chinese characters from Gitea API
|
||||||
|
# crash print() under the default Windows code page.
|
||||||
|
try:
|
||||||
|
sys.stdout.reconfigure(encoding='utf-8')
|
||||||
|
except Exception:
|
||||||
|
pass
|
||||||
|
|
||||||
def _load_gitea_config():
|
def _load_gitea_config():
|
||||||
"""Load Gitea URL, repo, and token from ~/.gitea/config.yaml or env vars."""
|
"""Load Gitea URL, repo, and token from ~/.gitea/config.yaml or env vars."""
|
||||||
config_path = os.path.expanduser("~/.gitea/config.yaml")
|
config_path = os.path.expanduser("~/.gitea/config.yaml")
|
||||||
@@ -173,7 +180,7 @@ def blocked_check():
|
|||||||
print(f"Checked {len(all_blocked)} blocked issue(s): still blocked.")
|
print(f"Checked {len(all_blocked)} blocked issue(s): still blocked.")
|
||||||
|
|
||||||
|
|
||||||
def get_issue(num):
|
def get_issue(num, with_comments=True):
|
||||||
i = _req("GET", f"/issues/{num}")
|
i = _req("GET", f"/issues/{num}")
|
||||||
print(f"## #{i['number']}: {i['title']}")
|
print(f"## #{i['number']}: {i['title']}")
|
||||||
print(f"State: {i['state']}")
|
print(f"State: {i['state']}")
|
||||||
@@ -181,6 +188,16 @@ def get_issue(num):
|
|||||||
print(f"Labels: {', '.join(labels) if labels else 'none'}")
|
print(f"Labels: {', '.join(labels) if labels else 'none'}")
|
||||||
print()
|
print()
|
||||||
print(i.get("body", "(no description)"))
|
print(i.get("body", "(no description)"))
|
||||||
|
if with_comments:
|
||||||
|
comments = _req_safe("GET", f"/issues/{num}/comments")
|
||||||
|
if comments:
|
||||||
|
print(f"\n--- Comments ({len(comments)}) ---")
|
||||||
|
for c in comments:
|
||||||
|
user = c.get("user", {}).get("login", "unknown")
|
||||||
|
created = c.get("created_at", "")[:16]
|
||||||
|
body = c.get("body", "")
|
||||||
|
print(f"\n[{user}] {created}")
|
||||||
|
print(body)
|
||||||
return i
|
return i
|
||||||
|
|
||||||
|
|
||||||
|
|||||||
@@ -19,15 +19,13 @@ source "$SCRIPT_DIR/_common.sh"
|
|||||||
setup_worktree "$GITEA_USER"
|
setup_worktree "$GITEA_USER"
|
||||||
|
|
||||||
# Cleanup worktree on exit (optional, comment out to keep for debugging)
|
# Cleanup worktree on exit (optional, comment out to keep for debugging)
|
||||||
trap 'cleanup_worktree "$GITEA_USER"' EXIT
|
trap 'cleanup_worktree' EXIT
|
||||||
|
|
||||||
banner "Dev"
|
banner "Dev"
|
||||||
require_token
|
require_token
|
||||||
|
|
||||||
AGENT_CONF="$_MAIN_REPO_DIR/.claude/agents/dev-agent.md"
|
|
||||||
launch_agent \
|
launch_agent \
|
||||||
"dev-agent" \
|
"dev-agent" \
|
||||||
"$AGENT_CONF" \
|
"$PROJECT_DIR/.claude/agents/dev-agent.md" \
|
||||||
"Dev-Agent" \
|
"Dev-Agent" \
|
||||||
"执行一次 Issue 巡检(单次任务,不要用 /loop):
|
"执行一次 Issue 巡检(单次任务,不要用 /loop):
|
||||||
1. python scripts/agent_poller.py --action list 列出所有打开的 Issue
|
1. python scripts/agent_poller.py --action list 列出所有打开的 Issue
|
||||||
|
|||||||
@@ -0,0 +1,22 @@
|
|||||||
|
#!/usr/bin/env bash
|
||||||
|
# Generic session 启动脚本 — 为人类用户提供 worktree 隔离
|
||||||
|
# 用法: bash scripts/start_generic.sh
|
||||||
|
# GITEA_USER 默认 pzhangzywl(人类用户),可通过环境变量覆盖
|
||||||
|
|
||||||
|
set -eu
|
||||||
|
|
||||||
|
export GITEA_USER="${GITEA_USER:-pzhangzywl}"
|
||||||
|
|
||||||
|
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
|
||||||
|
source "$SCRIPT_DIR/_common.sh"
|
||||||
|
|
||||||
|
setup_worktree "$GITEA_USER"
|
||||||
|
trap 'cleanup_worktree' EXIT
|
||||||
|
|
||||||
|
banner "Generic"
|
||||||
|
require_token
|
||||||
|
|
||||||
|
echo "工作目录: $PROJECT_DIR"
|
||||||
|
echo ""
|
||||||
|
cd "$PROJECT_DIR"
|
||||||
|
claude
|
||||||
@@ -19,15 +19,13 @@ source "$SCRIPT_DIR/_common.sh"
|
|||||||
setup_worktree "$GITEA_USER"
|
setup_worktree "$GITEA_USER"
|
||||||
|
|
||||||
# Cleanup worktree on exit (optional, comment out to keep for debugging)
|
# Cleanup worktree on exit (optional, comment out to keep for debugging)
|
||||||
trap 'cleanup_worktree "$GITEA_USER"' EXIT
|
trap 'cleanup_worktree' EXIT
|
||||||
|
|
||||||
banner "QE"
|
banner "QE"
|
||||||
require_token
|
require_token
|
||||||
|
|
||||||
AGENT_CONF="$_MAIN_REPO_DIR/.claude/agents/qe-agent.md"
|
|
||||||
launch_agent \
|
launch_agent \
|
||||||
"qe-agent" \
|
"qe-agent" \
|
||||||
"$AGENT_CONF" \
|
"$PROJECT_DIR/.claude/agents/qe-agent.md" \
|
||||||
"QE-Agent" \
|
"QE-Agent" \
|
||||||
"执行一次 Issue 巡检(单次任务,不要用 /loop):
|
"执行一次 Issue 巡检(单次任务,不要用 /loop):
|
||||||
1. python scripts/agent_poller.py --action list --labels test-code 检查 test-code Issue
|
1. python scripts/agent_poller.py --action list --labels test-code 检查 test-code Issue
|
||||||
|
|||||||
@@ -9,7 +9,7 @@ LLM configuration is read from secrets.yaml (searched in order):
|
|||||||
2. ~/.openclaw/config/secrets.yaml
|
2. ~/.openclaw/config/secrets.yaml
|
||||||
3. ~/.openclaw/workspace-document-analyzer/config/secrets.yaml
|
3. ~/.openclaw/workspace-document-analyzer/config/secrets.yaml
|
||||||
|
|
||||||
deepseek.apiKey / deepseek.baseUrl → text model (deepseek-v4-flash)
|
deepseek.apiKey / deepseek.baseUrl → text model (deepseek-v4-pro)
|
||||||
|
|
||||||
Environment variables:
|
Environment variables:
|
||||||
TEST_IR_PATH — path to IR JSON (default: output/final/ir_final.json)
|
TEST_IR_PATH — path to IR JSON (default: output/final/ir_final.json)
|
||||||
@@ -198,11 +198,11 @@ def parsed_data(parsed_path: str | None) -> dict | None:
|
|||||||
class _AcceptanceLLM:
|
class _AcceptanceLLM:
|
||||||
"""Thin LLM wrapper for acceptance tests.
|
"""Thin LLM wrapper for acceptance tests.
|
||||||
|
|
||||||
Uses deepseek-v4-flash for text (Layer C QE audit) via OpenAI-compatible API,
|
Uses deepseek-v4-pro for text (Layer C QE audit) via OpenAI-compatible API,
|
||||||
configured from ~/.openclaw/config/secrets.yaml.
|
configured from ~/.openclaw/config/secrets.yaml.
|
||||||
"""
|
"""
|
||||||
|
|
||||||
TEXT_MODEL = "deepseek-v4-flash"
|
TEXT_MODEL = "deepseek-v4-pro"
|
||||||
IMAGE_MODEL = "qwen3-vl-plus"
|
IMAGE_MODEL = "qwen3-vl-plus"
|
||||||
TIMEOUT = 180
|
TIMEOUT = 180
|
||||||
MAX_RETRIES = 3
|
MAX_RETRIES = 3
|
||||||
@@ -277,7 +277,7 @@ class _AcceptanceLLM:
|
|||||||
def llm_client():
|
def llm_client():
|
||||||
"""Create an LLM client for acceptance tests.
|
"""Create an LLM client for acceptance tests.
|
||||||
|
|
||||||
Uses deepseek-v4-flash for text (Layer C QE audit), configured from
|
Uses deepseek-v4-pro for text (Layer C QE audit), configured from
|
||||||
~/.openclaw/config/secrets.yaml deepseek section.
|
~/.openclaw/config/secrets.yaml deepseek section.
|
||||||
"""
|
"""
|
||||||
return _AcceptanceLLM()
|
return _AcceptanceLLM()
|
||||||
|
|||||||
Reference in New Issue
Block a user