v0.20.7 restructure source and platform workspaces

This commit is contained in:
Deep Research System
2026-05-07 21:13:21 +08:00
parent b3050260b2
commit a4adeb591f
172 changed files with 214 additions and 543 deletions
+28 -31
View File
@@ -5,17 +5,17 @@ secrets.env
*.pem *.pem
# ============ 字体(~140MB,通过 download-fonts.sh 获取)============ # ============ 字体(~140MB,通过 download-fonts.sh 获取)============
.opencode/templates/fonts/*.otf *.otf
.opencode/templates/fonts/*.ttf *.ttf
.opencode/templates/fonts/*.ttc *.ttc
.opencode/templates/fonts/*.woff* *.woff*
# ============ 研究产出物(大文件)============ # ============ 研究产出物(大文件)============
projects/*/phase4/*.pdf */projects/*/phase4/*.pdf
projects/*/phase4/*.docx */projects/*/phase4/*.docx
projects/*/phase4/figures/*.png */projects/*/phase4/figures/*.png
projects/*/phase2/evidence/*.cache */projects/*/phase2/evidence/*.cache
projects/*/**/*.tmp */projects/*/**/*.tmp
# ============ Python ============ # ============ Python ============
__pycache__/ __pycache__/
@@ -45,32 +45,28 @@ Thumbs.db
.idea/ .idea/
*.sublime-* *.sublime-*
# ============ OpenCode ============ # ============ Generated platform environments ============
.opencode/log/ /antigravity/
.opencode/cache/ /codex/
/opencode/
/claude-code/
/gemini-cli/
/platform_envs/
# ============ Codex adapter deployment target ============ # ============ Legacy root platform dirs ============
# v0.20 keeps Codex templates in codex_adapter_templates/ and deploys usable /.agent/
# files to $CODEX_HOME or ~/.codex via scripts/deploy_adapters.py. /.agents/
.codex/ /.codex/
/.opencode/
# ============ Generated / legacy platform adapter working dirs ============ /.claude/
# Antigravity native work should use tracked .agent/ only. These directories /.gemini/
# are generated for other tools or kept out of clean Antigravity workspaces.
.agents/
.opencode/
.claude/
.gemini/
# Rebuilt local platform workspaces. Generated by scripts/update_platform_envs.py.
platform_envs/
# Local platform permissions/settings are never source templates. # Local platform permissions/settings are never source templates.
platform_adapters/claude-code/settings.local.json src/platform_adapters/claude-code/settings.local.json
# ============ 归档(不纳入版本控制)============ # ============ 归档(不纳入版本控制)============
archive/* src/archive/*
!archive/.gitkeep !src/archive/.gitkeep
# ============ 临时 LaTeX / TeX 测试文件 ============ # ============ 临时 LaTeX / TeX 测试文件 ============
xetest.* xetest.*
@@ -86,4 +82,5 @@ _preamble.tex
# ============ 研究项目(实际数据,不纳入版本控制)============ # ============ 研究项目(实际数据,不纳入版本控制)============
# 如需备份,请用独立的私有仓库 # 如需备份,请用独立的私有仓库
projects/ src/projects/*
!src/projects/.gitkeep
+41 -497
View File
@@ -1,523 +1,67 @@
# Deep Research 系统 # Deep Research
> 生物医药行业的 AI 驱动深度研究流水线。v0.20 起以项目自有 Python core + skills + model profiles 为核心,以麦肯锡/德勤式方法论产出专业级研究报告(PDF + DOCX) 生物医药 Deep Research 仓库。根目录只保留入口说明和环境重建脚本;所有源文件放在 `src/`,各平台运行目录由脚本生成
**当前状态**:v0.20 重构启动。核心编排从平台 prompt 迁移到项目自有 Python runtime:模型选择、skills、任务卡并发、中文原生成稿都由 `scripts/dr.py``configs/models.yaml` 驱动;OpenCode/Codex/Claude Code/Antigravity/Gemini CLI 只作为表层入口。 ## Directory Layout
详见 `PLAN.md` 了解完整方案、版本记录与迭代路径。
--- ```text
## 快速开始
> 项目在 **macOS** 与 **Debian/Ubuntu** 上均通用。Python 包管理用 [**uv**](https://docs.astral.sh/uv/)Rust 写的 Python 包管理器,比 pip 快 10-100 倍,Astral 出品)。
### 1. 一键初始化(推荐)
```bash
bash scripts/setup.sh
```
这个脚本会自动:
- 检查并安装 **uv**(没有会交互询问,走官方 `curl -LsSf https://astral.sh/uv/install.sh` 脚本)
- 调用 `uv sync``pyproject.toml` 创建 `.venv/` 并装所有依赖(ReportLab、matplotlib、biopython 等)
- uv 会自动管理 Python 版本(>=3.10),不依赖系统 Python
- 检查并提示缺失的系统二进制(pandoc、opencode
**为什么选 uv 而不是 pip**
- macOS Homebrew Python、Debian 12+ 默认 PEP 668 保护,`pip install` 到系统会被拒
- uv 把 venv + 版本管理 + 依赖解析一次搞定,不需要手动 `python -m venv` + `source activate` + `pip install`
- `uv sync` 10-30 秒装完一整个科学计算栈(pip 要 2-5 分钟)
-`uv.lock` 锁定精确版本,mac 和 debian 装出来的环境完全一致
### 2. 安装系统二进制(setup.sh 不装这些)
```bash
# macOS
brew install pandoc
curl -fsSL https://opencode.ai/install | bash
# Debian/Ubuntu
sudo apt update && sudo apt install -y pandoc curl
curl -fsSL https://opencode.ai/install | bash
```
> uv 会帮你装 Python 本身,不需要再装 `python3-venv` / `python3-pip`。
### 3. 配置密钥
```bash
cp secrets.env.example secrets.env
# 编辑 secrets.env,填入:
# - ZENMUX_API_KEY(必填,格式 sk-ai-v1-xxx
# - TAVILY_API_KEY / BRAVE_API_KEY / EXA_API_KEY
# - NCBI_API_KEY(可选,PubMed 高频查询时用)
```
### 4. 下载中文字体
```bash
bash .opencode/templates/fonts/download-fonts.sh
```
下载思源宋体 + 思源黑体 + 霞鹜文楷(SIL OFL 许可,约 140MB)。
### 5. 验证 zenmux 端点与 cache(强烈推荐)
```bash
source scripts/activate.sh # 一键激活 venv + 加载 secrets
bash scripts/verify-zenmux.sh # 3 步自检
```
`verify-zenmux.sh` 会:
1. 测 OpenAI 兼容端点(Gemini 3.1 Pro Preview
2. 测 Anthropic 兼容端点(Claude Haiku 4.5
3. 测 Claude prompt cache 两次调用命中(Opus 4.7 cache 读取仅 0.5 USD/M tokens
若第 3 步 `cache_read_input_tokens` 始终为 0,参见 `AGENTS.md` §6.5 诊断步骤。
### 6. 日常使用
```bash
cd ~/Documents/Projects/deep_research
source scripts/activate.sh # 激活 venv + 载入 secrets
opencode # 启动 TUI
```
**退出 venv**`deactivate`
**跑单个 Python 脚本**(不用先激活):
```bash
uv run python scripts/build_report.py <slug> # 默认 ReportLab
uv run python scripts/build_report.py <slug> --engine quarto # Quarto/xelatex
```
网络不稳或 API 限流时,把 `--workers` 降到 `3``1` 即可断点续跑。
### 多平台兼容
v0.20 后,OpenCode/Codex/Claude Code/Gemini CLI 都是 Python core 的 surface adapter。各平台配置源放在非隐藏目录 `platform_adapters/`;本地运行环境由 `scripts/update_platform_envs.py` 重建到被忽略的 `platform_envs/`。Antigravity 走 native 模式,用自身模型配额按 phase 执行研究,Python core 只做辅助。详细调用方式见 `docs/platform-adapters.md`
- OpenCode:生成到 `platform_envs/opencode/.opencode/commands/*.md`,薄封装 Python CLI。
- Codex:使用 `AGENTS.md` + 部署到 `$CODEX_HOME` 的 adapter 文件,优先用 GPT 系列做代码/测试/审阅。
- Claude Code:生成到 `platform_envs/claude-code/.claude/skills/*/SKILL.md`,优先用 Opus/Sonnet 做访谈、结构讨论和中文风格审阅。
- Gemini CLI:生成到 `platform_envs/gemini-cli/.gemini/commands/dr/*.toml`,优先用 Gemini 做长上下文、多模态和替代框架审阅。
- Antigravity:打开 `platform_envs/antigravity`,使用其中的 `.agent/rules` + `.agent/skills` 指导其用 Antigravity 模型配额原生执行 Deep ResearchPython core 只做脚手架、校验、引用和出稿辅助。
更新/重建某个平台环境:
```bash
python scripts/update_platform_envs.py --platform antigravity
python scripts/update_platform_envs.py --platform all
```
Antigravity 推荐打开 clean workspace,不推荐直接打开完整多平台仓库。详见 `docs/antigravity-clean-workspace.md`
平台分支/环境隔离策略见 `docs/platform-branch-strategy.md`
部署 Codex adapter(不在仓库内创建 `.codex`):
```bash
uv run python scripts/deploy_adapters.py codex --force
```
Codex adapter 会写到 `$CODEX_HOME``~/.codex`;已有文件会在 `--force` 覆盖前生成 `.bak` 备份。adapter 默认面向自动化研究:workspace 可写、命令不逐次审批、实时 web search 与脚本网络访问开启;Tavily / Brave / Exa MCP 会默认启用但不设为必需服务。
安全默认:部署脚本不会复制 `config.toml`,避免覆盖用户级 Codex 配置。只有明确需要安装本项目 bundled profile 时,才使用 `--include-config`
部署 Antigravity workspace adapter(从 `platform_adapters/antigravity` 写到目标工作区 `.agent/`,默认不覆盖已有文件):
```bash
uv run python scripts/deploy_adapters.py antigravity --target /path/to/workspace --dry-run
uv run python scripts/deploy_adapters.py antigravity --target /path/to/workspace
```
Antigravity 部署不会修改全局设置。若目标工作区已有同名 agents/rule/skill/workflow 文件,默认跳过;只有显式加 `--force` 时才会先生成 `.bak` 备份再覆盖。Antigravity 分层为:`.agent/agents.md` 定义角色,`.agent/rules` 放强约束,`.agent/skills` 放能力,`.agent/workflows` 放流程。关键 artifact 包括 `search_log.jsonl``sources.jsonl``claims_ledger.jsonl``coverage_matrix.md``continuation_state.json`
部署到新环境后自检:
```bash
uv run python scripts/deploy_check.py
uv run python scripts/deploy_check.py --repair --force
```
运行平台无关 Python core
```bash
uv run python scripts/dr.py init "研究主题" --slug <slug> --method mckinsey_market
uv run python scripts/dr.py frame <slug>
uv run python scripts/dr.py run <slug-or-topic> --method gmp_gap_assessment
uv run python scripts/dr.py research <slug> --workers 6
uv run python scripts/dr.py research <slug> --workers 6 --execute-packets
uv run python scripts/dr.py research <slug> --workers 6 --execute-packets --allow-search-fallback
uv run python scripts/dr.py research <slug> --workers 6 --build-briefs
uv run python scripts/dr.py research <slug> --workers 6 --assemble-chapters
uv run python scripts/dr.py review <slug>
uv run python scripts/dr.py finalize <slug>
```
OpenCode/Codex/Claude Code/Gemini CLI adapter 只包装这些 CLI,不再承担核心调度或模型选择。Antigravity 是例外:它优先按生成环境中的 `.agent/skills/antigravity-surface-adapter` 用自身模型配额执行研究,避免主流程重新落回 ZenMux。
模型与搜索 API 选择见:
- `docs/model-playbook.md`
- `docs/search-playbook.md`
模型预设配置文件:
- `configs/models.yaml`(统一预设,支持 `simple / medium / premium / cn_heavy / codex_native`
推荐时机:在 `/dr-init` 访谈阶段就确定 `model_profile`。v0.20 后模型选择优先在 Python runtime 中解析,adapter agent 文件只是兼容层。
命令行查看解析后的模型映射:
```bash
uv run python scripts/dr.py models
uv run python scripts/dr.py models --list
uv run python scripts/dr.py models --profile premium
uv run python scripts/dr.py models --profile medium --model-override dr_verifier=zenmux/openai/gpt-5.4
uv run python scripts/dr.py skills validate
uv run python scripts/dr.py skills sync
uv run python scripts/dr.py methods list
uv run python scripts/dr.py methods show gmp_gap_assessment
# apply profile to agent files
uv run python scripts/dr.py apply-models --profile medium --target both --dry-run
uv run python scripts/dr.py apply-models --profile medium --target both
```
Sprint 5 回归检查(一键):
```bash
uv run python scripts/sprint5_regression.py <slug>
```
统一搜索入口:
```bash
uv run python scripts/search.py "dual-target RNAi 2024" --route scholar --year-low 2023
uv run python scripts/search.py "FDA warning letter CAPA deviation change control pharmaceutical" --route evidence --json --trace
uv run python scripts/search.py "dual-target siRNA GalNAc" --route patents
```
v0.20 搜索分工:`evidence` 用 Exa highlights 做受控候选证据发现;`scholar/patents/news/fda` 保留专用路由;`general` 只做宽泛发现和兜底;Tavily Research 作为 Phase1 初扫、薄弱章节补证据和 Phase3 回炉工具,结果必须存盘、评分、去重后再进入 evidence packet。Antigravity 原生流程可优先使用 `paperfoot/search-cli``search` 命令作为统一检索前台,但所有结论仍必须回到本项目 Tier/source-quality 体系。
---
## 项目结构
```
deep_research/ deep_research/
├── PLAN.md # 完整方案(中断续接从此读起) README.md
├── AGENTS.md # 跨平台研究方法论与运行规则 setup.sh
├── README.md # 本文件 src/ # canonical source: scripts, skills, docs, configs, platform adapters
├── secrets.env.example # 密钥模板 codex/ # generated Codex workspace
├── secrets.env # 你的密钥(gitignore antigravity/ # generated Antigravity workspace
opencode/ # generated OpenCode workspace
├── .opencode/ claude-code/ # generated Claude Code workspace
│ ├── opencode.json # MCP 配置 + 权限 gemini-cli/ # generated Gemini CLI workspace
│ ├── agents/ # Agent 定义
│ │ ├── dr-plan.md # [MVP] 框架规划师 Opus 0.7
│ │ └── dr-pm.md # [MVP] 项目经理 Sonnet 0.2
│ ├── skills/ # 可复用技能
│ │ ├── search-strategy/ # [MVP] 检索策略总纲
│ │ ├── source-quality/ # [MVP] 信源评级
│ │ ├── length-budget/ # [MVP] 字数预算
│ │ └── pdf-reportlab/ # [MVP] PDF 模板使用
│ ├── commands/
│ │ ├── dr-init.md # [MVP] /dr-init
│ │ └── dr-frame.md # [MVP] /dr-frame
│ └── templates/
│ ├── report-template.py # [MVP] ReportLab PDF 生成器
│ └── fonts/
│ ├── download-fonts.sh
│ └── README.md
├── projects/ # 每个研究一个子目录
│ └── <slug>/
│ ├── manifest.json
│ ├── phase1/
│ ├── phase2/
│ ├── phase3/
│ └── phase4/
└── archive/ # 完成项目归档
``` ```
--- Generated platform directories are ignored by Git. Rebuild them whenever source templates change.
## 关键设计要点 ## Setup
### 1. Python core 防止上下文污染 Initialize all platform workspaces from local `src/`:
OpenCode/Codex/Claude Code/Antigravity/Gemini CLI 的 subagent 或 agent thread 都可能把大上下文带入子会话。v0.20 起,默认并发由 Python runtime 的任务卡执行层控制:先生成 `phase1/research_brief.json` 固化研究意图和任务切分,再生成 `phase2/task_cards.json`,按任务卡产出 `phase2/packets/*.json`,经 `phase2/compressed_findings/*.json` 收束后再组装中文章节。平台 agent 只负责调用 CLI 和展示状态。
### 2. 信源分级(Tier 1-4 + 黑名单)
详见 `AGENTS.md` §4 与 `skills/source-quality/SKILL.md`
- **Tier 1**PubMed、顶刊、FDA/NMPA 监管、ClinicalTrials、专利
- **Tier 2**:权威咨询、系统综述、行业协会
- **Tier 3**:预印本、券商研报、会议摘要(需 Tier 1-2 支撑)
- **Tier 4**:通用搜索(仅做发现入口,不做证据)
- **黑名单**:自媒体、撤稿论文、软文
### 3. 字数硬性要求
| 类型 | 最小字数 |
|---|---|
| 综述 | 10,000 |
| 研究 | 30,000 |
| 投资 | 20,000 |
| 管理 | 15,000 |
Phase 1 分配章节配额,Phase 2 自检,不足返工。见 `skills/length-budget/SKILL.md`
### 4. 中文 PDF 双引擎
`build_report.py` 现在支持两套 PDF 引擎,按需选择:
#### 引擎 AReportLab(默认,无额外依赖)
```bash ```bash
uv run python scripts/build_report.py <slug> ./setup.sh init all
``` ```
- 字体:思源宋 + 思源黑 + 霞鹜文楷(全 SIL OFL,可商用嵌入) Initialize one platform:
- 样式:集中在 `build_styles()`,所有字号行距单点维护
- 速度:30,000 字 3-5 秒出稿
- 局限:超宽表格(≥4 列且含长文本)需借助列宽 patch 或改为 bullet list 格式
#### 引擎 BQuarto / xelatex`--engine quarto`,推荐用于宽表报告)
```bash ```bash
uv run python scripts/build_report.py <slug> --engine quarto ./setup.sh init antigravity
./setup.sh init codex
./setup.sh init opencode
./setup.sh init claude-code
./setup.sh init gemini-cli
``` ```
- 排版引擎:xelatexTeX Live / TinyTeX),LaTeX 级排版质量 Pull latest changes and rebuild:
- 字体:同样使用思源宋 + 思源黑,通过 fontspec 加载
- 宽表支持:超宽表通过 `longtable` + `tbl-colwidths` 精确指定列宽比例,不溢出
- 横向页面:通过 `{.landscape}` div 包裹超宽表,自动插入 `pdflscape` 代码(注意:101 行以上的 landscape longtable 可能触发 TeX `param_size` 上限,建议拆成 ≤20 行的子表块)
- 图表:暂不嵌入 matplotlib 图表(使用文字描述代替)
**安装 Quarto 引擎**(一次性,系统级):
```bash ```bash
# 1. 安装 Quarto CLI ./setup.sh update all
# 下载页:https://github.com/quarto-dev/quarto-cli/releases/latest ./setup.sh update antigravity
# Linux 选 .deb 安装包,macOS 选 .pkg
# 2. 安装 TinyTeXQuarto 内置命令)
quarto install tinytex
# 3. 安装中文 LaTeX 支持包
~/.TinyTeX/bin/x86_64-linux/tlmgr install ctex xecjk cjk xetex
# macOS 路径通常为:~/.TinyTeX/bin/universal-darwin/tlmgr
# 4. 注册思源字体到 fontconfig
# (先确认字体已下载:bash .opencode/templates/fonts/download-fonts.sh
mkdir -p ~/.fonts
cp .opencode/templates/fonts/*.otf ~/.fonts/
cp .opencode/templates/fonts/*.ttf ~/.fonts/
cp .opencode/templates/fonts/ttf/*.ttf ~/.fonts/
fc-cache -fv ~/.fonts
# 5. 验证
quarto --version # 应输出 1.x.x
fc-list | grep "Source Han" # 应看到思源字体条目
``` ```
**两引擎对比** After rebuilding, open the generated platform directory in the matching tool:
| 指标 | ReportLab | Quarto/xelatex | - Codex: `deep_research/codex`
|------|-----------|----------------| - Antigravity: `deep_research/antigravity`
| 安装复杂度 | 无额外依赖 | 需安装 Quarto + TinyTeX | - OpenCode: `deep_research/opencode`
| 渲染速度 | 3-5 秒 | 30-90 秒(LaTeX 编译) | - Claude Code: `deep_research/claude-code`
| 宽表格处理 | 需 workaround | longtable 原生支持 | - Gemini CLI: `deep_research/gemini-cli`
| 横向页面 | 不支持 | 支持(≤20 行/块) |
| 字体嵌入 | OTF 直接嵌入 | fontspec 系统字体 |
| 输出体积 | ~1.2 MB/100页 | ~0.9 MB/100页 |
| 目录生成 | 自定义实现 | LaTeX 自动 \tableofcontents |
### 5. zenmux 双 providerClaude cache 关键) ## Source
Deep Research 大量使用**相同的长 system prompt + skill 内容**连续调用 Claude,如果 prompt cache 没生效,Opus/Sonnet 成本会翻 5-10 倍。所以本项目: Shared skills and research rules should live in `src/` and be reused by platform adapters. Avoid duplicating method, source-quality, citation, evidence, and output-hygiene rules inside platform-specific files unless the platform requires a thin wrapper.
- **Claude 系列** → `zenmux-anthropic/claude-opus-4.7`(走 `https://zenmux.ai/api/anthropic``@ai-sdk/anthropic` 原生支持 `cache_control` Key source locations:
- **非 Claude 系列** → `zenmux/google/gemini-3.1-pro-preview` 等(走 `https://zenmux.ai/api/v1`,隐式缓存自动生效)
Opus 4.7 cache 读取价格 0.5 USD/M tokens(对比输入 25 USD/M,节省 98%)。 - `src/scripts/`
- `src/configs/`
- `src/skills/`
- `src/platform_adapters/`
- `src/docs/`
- `src/PLAN.md`
**验证 cache 是否生效** Platform-specific behavior is documented in each generated workspace README and in `src/docs/platform-branch-strategy.md`.
1. 在 https://zenmux.ai/settings/logs 打开 API Call Logging
2. 运行一个真实 Claude/ZenMux 调用链路,例如 `uv run python scripts/dr.py research <slug> --execute-packets`
3. 第 2 次的 `cache_read_input_tokens` 字段应 > 0
4. 若始终为 0,检查 `configs/models.yaml` 中对应 role 是否走 `zenmux-anthropic/...`
---
## 已知待办
### 用户侧
1. [x] ~~在 zenmux 后台确认模型 slug,更新 `AGENTS.md` §6 映射表~~(已在 v0.2 完成,使用双 provider 架构)
2. [ ] 填写 `secrets.env`
3. [ ] 运行 `download-fonts.sh` 下载字体
4. [ ] 用一个小主题(如"5000 字 PD-1 综述")跑通 MVP 流水线
### 系统侧(下一阶段)
- [x] Python core `init/frame/research/review/finalize/status` 骨架
- [x] OpenCode/Codex/Claude Code/Gemini CLI wrapper
- [x] Antigravity 专用 workspace rule + skill,按 Antigravity 模型配额原生执行研究
- [x] 用户资料导入基础能力:PDF 文本抽取 + FireRed OCR 扫描件识别 + phase0 落盘
- [ ] 用户资料导入增强:DOCX / PPTX / 表格抽取 / 版面结构化
- [ ] PDF reporting 包继续拆分:字体、宽表、引用、渲染验证
---
## 排错
### 字体下载失败
```bash
# 检查网络访问 GitHub
curl -I https://github.com
# 用镜像手动下载(见 .opencode/templates/fonts/README.md
```
### MCP Server 启动失败
```bash
# 检查 npx 可用
which npx
# 验证 MCP server 可独立运行
npx -y tavily-mcp@latest
```
### 平台 agent 看起来没有真正并发
v0.20 不再用平台 subagent 作为默认并发机制。请检查 Python core 产物:
```bash
uv run python scripts/dr.py status <slug>
ls projects/<slug>/phase2/task_cards.json
ls projects/<slug>/phase2/packets
ls projects/<slug>/phase2/chapter_briefs
```
### ReportLab PDF 中文乱码
```bash
# 确认字体已注册(venv 里)
source scripts/activate.sh
ls .opencode/templates/fonts/*.otf | wc -l # 应为 6+
# 手动测试
python .opencode/templates/report-template.py --help
```
### Quarto PDF 生成失败
**字体找不到(`Could not resolve font "Source Han Serif CN/I"`**
CJK 字体没有斜体变体,fontspec 默认会找 `/I` 导致报错。`build_report.py --engine quarto` 已通过 `mainfontoptions: [ItalicFont=...]` 自动绕开,无需手动处理。若自行编写 `.qmd`,需在 YAML 里加:
```yaml
format:
pdf:
pdf-engine: xelatex
CJKmainfont: "Source Han Serif CN"
mainfontoptions:
- BoldFont=Source Han Serif CN
- ItalicFont=Source Han Serif CN
- BoldItalicFont=Source Han Serif CN
```
**`tlmgr` 找不到**
TinyTeX 不在系统 PATH,用完整路径:
```bash
~/.TinyTeX/bin/x86_64-linux/tlmgr install <package> # Linux
~/.TinyTeX/bin/universal-darwin/tlmgr install <package> # macOS
```
**`TeX capacity exceeded [parameter stack size]`**landscape 大表):
pdflscape 的 `\LS@makefcolumn` 在 101 行以上的 longtable 里递归过深,耗尽 TeX 的 `param_size`。解决方法:把超大表拆成每块 ≤20 行的子表,每块都包在 `{.landscape}` div 里:
```markdown
::: {.landscape}
| 列1 | 列2 | ... |
|---|---|---|
| 第1-20行 | ... |
:::
::: {.landscape}
| 列1 | 列2 | ... |
|---|---|---|
| 第21-40行 | ... |
:::
```
若使用 `build_report.py --engine quarto`,可通过传入预处理好的 `.md`(宽表已拆块)来避免此问题。
### uv 安装后找不到
uv 官方脚本把 uv 装到 `~/.local/bin/`。若终端里 `which uv` 找不到:
```bash
export PATH="$HOME/.local/bin:$PATH"
# 永久生效:
echo 'export PATH="$HOME/.local/bin:$PATH"' >> ~/.zshrc # macOS zsh
echo 'export PATH="$HOME/.local/bin:$PATH"' >> ~/.bashrc # debian bash
```
### Python 依赖安装报 externally-managed-environment
不要用系统 pip!改用 uv(本项目标准方案):
```bash
bash scripts/setup.sh # uv sync 会自建 .venv 并装依赖
```
### uv sync 很慢 / 下载超时
国内网络下,可用镜像加速:
```bash
export UV_INDEX_URL=https://pypi.tuna.tsinghua.edu.cn/simple
# 或
export UV_INDEX_URL=https://mirrors.aliyun.com/pypi/simple
uv sync
```
### 每次开终端都要手动 source
`source scripts/activate.sh` 一次搞定。或装 [direnv](https://direnv.net/)
```bash
# macOS
brew install direnv
# Debian
sudo apt install direnv
# 都装完后在项目根目录
echo 'source scripts/activate.sh' > .envrc
direnv allow
```
之后 `cd` 进项目目录就会自动激活,不用再想。
---
## 参考链接
- OpenCode 文档:https://opencode.ai/docs
- Agent 配置:https://opencode.ai/docs/agents
- 跨平台调用:`docs/platform-adapters.md`
- Skill 配置:https://opencode.ai/docs/skills
- MCP Servershttps://opencode.ai/docs/mcp-servers
- ReportLab 文档:https://docs.reportlab.com
- Quarto 文档:https://quarto.org/docs/output-formats/pdf-basics.html
- Quarto PDF 引擎:https://quarto.org/docs/output-formats/pdf-engine.html
- Quarto 表格文档:https://quarto.org/docs/authoring/tables.html
- 思源字体:https://github.com/adobe-fonts
- 霞鹜文楷:https://github.com/lxgw/LxgwWenKai
---
## 变更记录
- **v0.1** (2026-04-20) — MVP 路径 2 完成:dr-plan + dr-pm 两主 agent、4 个核心 skill、2 个命令、ReportLab 模板基础版、字体下载脚本
- **v0.2** (2026-04-20) — 双 provider 架构(zenmux-anthropic + zenmux),解决 Claude prompt cache 生效问题
- **v0.3** (2026-04-20) — 修正 v0.2 模型名(回到 Opus 4.7 / Sonnet 4.6 / Gemini 3.1 Pro / GPT-5.4 Pro 等真实 slug);改 venv + requirements.txt 跨平台方案(macOS + Debian);新增 `scripts/setup.sh``scripts/activate.sh`
- **v0.13** (2026-05-02) — `build_report.py` 新增 `--engine quarto` 选项:Quarto 1.9 + xelatex 引擎,解决 ReportLab 超宽表格渲染 bug`negative availWidth`/`NoneType` 问题);`report-template.py` 同步修复(`render_table_blocks` 分块 + 等宽列强制分配);README 补充双引擎安装指南与排错
- **v0.20** (2026-05-05) — Skill-driven Python core 重构启动:新增 `scripts/runtime/*``scripts/dr.py init/frame/run/research/review/skills`、task-type 模型映射、中文原生 finalize 默认路径和 `scripts/reporting/*` 报告模块;OpenCode/Codex/Claude Code/Gemini CLI 命令降级为 Python CLI wrapper。
- **v0.20.1** (2026-05-07) — Antigravity-native adapter:按 Gemini Flash 管理、Opus/Gemini 分 phase 执行研究;Python core 退为辅助工具。
`PLAN.md` §12 了解完整变更历史。
Executable
+45
View File
@@ -0,0 +1,45 @@
#!/usr/bin/env bash
set -euo pipefail
ROOT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
PYTHON_BIN="${PYTHON_BIN:-python3}"
usage() {
cat <<'EOF'
Usage:
./setup.sh init [platform|all]
./setup.sh update [platform|all]
Platforms:
antigravity
codex
opencode
claude-code
gemini-cli
all
Commands:
init Rebuild generated platform environments from src/ without pulling.
update Run git pull --ff-only, then rebuild generated platform environments.
EOF
}
cmd="${1:-}"
platform="${2:-all}"
case "$cmd" in
init)
"$PYTHON_BIN" "$ROOT_DIR/src/scripts/update_platform_envs.py" --platform "$platform" --skip-pull
;;
update)
"$PYTHON_BIN" "$ROOT_DIR/src/scripts/update_platform_envs.py" --platform "$platform"
;;
-h|--help|help|"")
usage
;;
*)
echo "Unknown command: $cmd" >&2
usage >&2
exit 2
;;
esac
View File
View File
View File
View File
@@ -0,0 +1,23 @@
# Antigravity Environment
Open this directory in Antigravity.
Antigravity is the only platform in this project that should run Deep Research natively with platform model quota. Use Gemini 3 Flash only as the surface manager, then switch models by phase as instructed in `.agent/workflows/deep-research-native.md`.
Start by loading:
- `AGENTS.md`
- `GEMINI.md`
- `.agent/agents.md`
- `.agent/rules/deep-research-antigravity.md`
- `.agent/skills/antigravity-surface-adapter/SKILL.md`
- `.agent/skills/method-selection/SKILL.md`
- `.agent/skills/research-quality-gates/SKILL.md`
- `.agent/workflows/deep-research-native.md`
Do not open the parent repository for production research. Rebuild this environment from the parent repository with:
```bash
./setup.sh init antigravity
./setup.sh update antigravity
```
@@ -0,0 +1,12 @@
# Claude Code Environment
Open this directory in Claude Code.
Claude Code is best used for interview-style framing, long-form Chinese editorial review, final wording quality, and careful reasoning around report structure. Shared research rules and skills still come from the generated workspace files, not from platform-specific rewrites.
Rebuild from the parent repository with:
```bash
./setup.sh init claude-code
./setup.sh update claude-code
```
+14
View File
@@ -0,0 +1,14 @@
# Codex Environment
Open this directory in Codex.
Codex is best used for repository work: scripts, schemas, tests, deployment logic, and code review. Deep Research model routing should normally use the Python core unless the user explicitly asks for manual Codex reasoning.
This environment contains a local `.codex/` adapter generated from `src/platform_adapters/codex` plus shared skills copied into `.codex/skills`.
Rebuild from the parent repository with:
```bash
./setup.sh init codex
./setup.sh update codex
```
@@ -0,0 +1,12 @@
# Gemini CLI Environment
Open this directory for Gemini CLI workflows.
Gemini CLI is suitable for long-context review, multimodal material checks, and alternative framework evaluation. Treat Gemini-family factual claims as high risk unless search receipts, source IDs, and claim-ledger rows exist.
Rebuild from the parent repository with:
```bash
./setup.sh init gemini-cli
./setup.sh update gemini-cli
```
+12
View File
@@ -0,0 +1,12 @@
# OpenCode Environment
Open this directory in OpenCode.
OpenCode is a thin surface adapter around the Python core. Use the `.opencode/commands` wrappers to call `scripts/dr.py`; do not let platform agents reimplement Phase 2 worker orchestration.
Rebuild from the parent repository with:
```bash
./setup.sh init opencode
./setup.sh update opencode
```

Some files were not shown because too many files have changed in this diff Show More