把 expert/raw/ 下新增素材,
抽到 expert/wiki/ 知识库

分批 — greedy bin-packing 按文件 size

# 新增 ≥ 30 → 4 路;15-30 → 2-3 路;< 15 → 单 agent
# 路数定了后用 greedy 算法把 N 个文件均匀塞 4 个 batch:
cd /Users/john/InvesResearch && python3 -c "
import json
from pathlib import Path
ext = Path('expert/wiki/_extracted')
PREFIXES = ('2科普视频文案__', '3其余素材__')   # 改这里适配本轮新增
new = []
for prefix in PREFIXES:
    for p in ext.glob(prefix + '*.md'):
        new.append((p.stat().st_size, p.name))
new.sort(reverse=True)
N_BATCHES = 4
batches = [[] for _ in range(N_BATCHES)]
sizes = [0]*N_BATCHES
for sz, name in new:
    i = sizes.index(min(sizes))
    batches[i].append(name); sizes[i] += sz
ROUND = 2; START = 8  # round 1 = 1..7;round 2 = 8..11 接序
out = {f'agent-{i+START}-round{ROUND}': batches[i] for i in range(N_BATCHES)}
Path('/tmp/wiki_round_batches.json').write_text(json.dumps(out, ensure_ascii=False, indent=2), encoding='utf-8')
for i, b in enumerate(batches):
    print(f'BATCH {i+1}: {len(b)} files, {sizes[i]/1024:.1f} KB')
"

每个 sub-agent 的提示词骨架

你是 InvesResearch 商业航天 wiki 增量入库的 round-{ROUND} sub-agent #{N}。
从新增料里抽 entities/people/concepts/reports/timeline,merge.py 会合并到 wiki/。

## 工作目录
- 仓库根:/Users/john/InvesResearch
- 源:expert/wiki/_extracted/*.md(只读你 batch 里的 N 个文件)
- 输出(已建好):
  - expert/wiki/_drafts/agent-{N}-round{ROUND}/{entities,people,concepts,reports}/
  - expert/wiki/_drafts/agent-{N}-round{ROUND}/timeline.md

## 你的 batch (N 文件, ~270KB)
{逐行列文件名 + 一句话主题提示}

## 输出契约
见 expert/wiki/_drafts/_TEMPLATE.md(已有,不重贴)
frontmatter:
- entities: type / country / aliases / status / sources
- people:   role / affiliation / country / aliases / sources
- concepts: category / aliases / sources
- reports:  title / publish_year / publisher / report_type / source_files

## Slug 规则 — 关键(决定 merge.py 能不能合)
先 Read 4 个现有 slug 清单 dedup:
- expert/wiki/_drafts/_round{ROUND}_existing_entities.txt
- expert/wiki/_drafts/_round{ROUND}_existing_people.txt
- expert/wiki/_drafts/_round{ROUND}_existing_concepts.txt
- expert/wiki/_drafts/_round{ROUND}_existing_reports.txt

- 已有 slug → 同名复用,merge.py 会 union frontmatter + 把你的正文追加为 ## 补充视角 — agent-{N}-round{ROUND}
- 新 slug 用 latin / pinyin(spacex / starlink / elon-musk / direct-to-cell),不要中文 slug(unicode normalize 不一致会撞坑)

## 质量纪律
1. 不编造 — 事实必须在源文本里找到出处
2. 简洁(每条 30-80 行 markdown)
3. sources 只列你 batch 里出现过的文件名
4. 同一实体一条;不要把同物分塞 entities + concepts
5. 关联 1-2 个 [[...]] wikilink
6. reports 必出 N 条(每个源 .md 一条 reports)
   - 科普视频文案 → report_type: 历史/科普,publish_year 留空
   - 长 PDF 行业报告 → report_type: 行业研究,有 publish_year 就填
   - 公司尽调 PDF → report_type: 公司尽调,敏感信息降级(估值数走区间)
7. 优先重要的;一带而过的不要写
8. timeline.md 按时间倒序 - YYYY-MM[-DD]: 简述,来源 `文件名.md`

## 完成后
短汇报"产出 N entities / M people / K concepts / R reports / T timeline 条"。不 commit。

# 0 · 先看新增了什么
ls expert/raw/<new-folder>/ | head -20

# 1 · Phase 1 抽取(增量,自动跳过已抽过的)
python3 expert/wiki/_scripts/extract.py

# 2 · 删空 PDF 输出(图扫描类抽不出文字)+ 打 fail 补丁
find expert/wiki/_extracted -name "*.md" -size -500c -delete
# 见 §踩坑 #2 的 JSON patch

# 3 · 切批 + 准备 dedup 助手 + 建空草稿目录
# 详见 §3 的 greedy bin-pack 脚本
ls expert/wiki/entities/  > expert/wiki/_drafts/_round${ROUND}_existing_entities.txt
ls expert/wiki/people/    > expert/wiki/_drafts/_round${ROUND}_existing_people.txt
ls expert/wiki/concepts/  > expert/wiki/_drafts/_round${ROUND}_existing_concepts.txt
ls expert/wiki/reports/   > expert/wiki/_drafts/_round${ROUND}_existing_reports.txt
for n in $(seq $START $((START+N-1))); do
  mkdir -p "expert/wiki/_drafts/agent-${n}-round${ROUND}"/{entities,people,concepts,reports}
done

# 4 · 派 N 路 sub-agent (一个 message N 个 Agent 调用,见 §3)
⚠️ 不是 bash!切到 Claude Code Agent 工具,run_in_background: true

# 5 · 等所有 sub-agent 完成后,验收草稿数量
for n in $(seq $START $((START+N-1))); do
  echo "== agent-${n}-round${ROUND} =="
  for sub in entities people concepts reports; do
    echo "  ${sub}: $(ls expert/wiki/_drafts/agent-${n}-round${ROUND}/${sub}/ 2>/dev/null | wc -l)"
  done
  echo "  timeline: $(wc -l < expert/wiki/_drafts/agent-${n}-round${ROUND}/timeline.md)"
done

# 6 · Phase 3 merge
python3 expert/wiki/_scripts/merge.py

# 7 · Phase 4 索引
python3 expert/wiki/_scripts/build_index.py

# 8 · Phase 5 HTML(自包含浏览器 + 处理进度 + 文档导航)
python3 expert/wiki/_scripts/build_html.py
python3 expert/wiki/_scripts/build_processing_status.py
python3 expert/wiki/_scripts/build_docs.py
# 可选:
python3 expert/wiki/_scripts/redact_check.py          # PII 审计
python3 expert/wiki/_scripts/build_redact_html.py     # PII 审计可视化

# 9 · 验收
echo "entities=$(ls expert/wiki/entities/ | wc -l)"
echo "people=$(ls expert/wiki/people/ | wc -l)"
echo "concepts=$(ls expert/wiki/concepts/ | wc -l)"
echo "reports=$(ls expert/wiki/reports/ | wc -l)"
echo "timeline lines=$(wc -l < expert/wiki/timeline/all.md)"
echo "index.html size=$(wc -c < expert/wiki/index.html)"

# 10 · 本地抽检
cd expert/wiki && python3 -m http.server 8765
# → http://localhost:8765/index.html
# → http://localhost:8765/processing_status.html