huangkiki/dailypaper-skills

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

Executable file; content unavailable

#!/usr/bin/env python3

#!/usr/bin/env python3

Executable file; content unavailable

#!/usr/bin/env python3

Executable file; content unavailable

#!/usr/bin/env python3

Executable file; content unavailable

#!/usr/bin/env python3

Executable file; content unavailable

#!/usr/bin/env python3

Executable file; content unavailable

#!/usr/bin/env python3

Executable file; content unavailable

#!/usr/bin/env python3

Executable file; content unavailable

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

- 合并优先级（脚本内部处理）：   - figure_url: HTML curl   - affiliations: PDF > HTML > abs fallback > Phase 1 data

- 合并优先级（脚本内部处理）：   - figure_url: HTML curl   - affiliations: PDF > HTML > abs fallback > Phase 1 data

- figure_url: HTML curl   - affiliations: PDF > HTML > abs fallback > Phase 1 data   - authors: HTML > abs fallback > Phase 1 data

- figure_url: HTML curl   - affiliations: PDF > HTML > abs fallback > Phase 1 data   - authors: HTML > abs fallback > Phase 1 data

#!/usr/bin/env python3

### N. 论文标题 - **链接**: [arXiv](https://arxiv.org/abs/XXXX) | [PDF](https://arxiv.org/pdf/XXXX)

### N. 论文标题 - **链接**: [arXiv](https://arxiv.org/abs/XXXX) | [PDF](https://arxiv.org/pdf/XXXX)

- **链接**: [arXiv](https://arxiv.org/abs/XXXX) | [PDF](https://arxiv.org/pdf/XXXX)

- **链接**: [arXiv](https://arxiv.org/abs/XXXX) | [PDF](https://arxiv.org/pdf/XXXX)

- **作者**: 完整作者列表（优先使用富化的 authors 字段，其次用原始 authors 字段） - **机构**: 从富化的 affiliations 字段获取，列出所有机构。如果 affiliations 为空，再检查原始 affiliations 字段。都没有则写"未知" - **链接**: [arXiv](https://arxiv.org/abs/XXXX) | [PDF](https://arxiv.org/pdf/XXXX)

- **作者**: 完整作者列表（优先使用富化的 authors 字段，其次用原始 authors 字段） - **机构**: 从富化的 affiliations 字段获取，列出所有机构。如果 affiliations 为空，再检查原始 affiliations 字段。都没有则写"未知" - **链接**: [arXiv](https://arxiv.org/abs/XXXX) | [PDF](https://arxiv.org/pdf/XXXX)

- **机构**: 从富化的 affiliations 字段获取，列出所有机构。如果 affiliations 为空，再检查原始 affiliations 字段。都没有则写"未知" - **链接**: [arXiv](https://arxiv.org/abs/XXXX) | [PDF](https://arxiv.org/pdf/XXXX) - **来源**: {见下方来源格式}

- **机构**: 从富化的 affiliations 字段获取，列出所有机构。如果 affiliations 为空，再检查原始 affiliations 字段。都没有则写"未知" - **链接**: [arXiv](https://arxiv.org/abs/XXXX) | [PDF](https://arxiv.org/pdf/XXXX) - **来源**: {见下方来源格式}

- **链接**: [arXiv](https://arxiv.org/abs/XXXX) | [PDF](https://arxiv.org/pdf/XXXX)

- **链接**: [arXiv](https://arxiv.org/abs/XXXX) | [PDF](https://arxiv.org/pdf/XXXX)

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

For each external image link ![...](https://...):

For each external image link ![...](https://...):

_DUPE_PATH_RE = re.compile(     r"(https://arxiv\.org/html/)(\d{4}\.\d{4,5}v\d+)/\2/"

_DUPE_PATH_RE = re.compile(     r"(https://arxiv\.org/html/)(\d{4}\.\d{4,5}v\d+)/\2/"

r"(https://arxiv\.org/html/)(\d{4}\.\d{4,5}v\d+)/\2/"

r"(https://arxiv\.org/html/)(\d{4}\.\d{4,5}v\d+)/\2/"

proc = await asyncio.create_subprocess_exec(                     "curl", "-sL", "--max-time", "30",                     "-o", pdf_path, f"https://arxiv.org/pdf/{arxiv_id}.pdf",

proc = await asyncio.create_subprocess_exec(                     "curl", "-sL", "--max-time", "30",                     "-o", pdf_path, f"https://arxiv.org/pdf/{arxiv_id}.pdf",

"curl", "-sL", "--max-time", "30",                     "-o", pdf_path, f"https://arxiv.org/pdf/{arxiv_id}.pdf",                     stdout=asyncio.subprocess.DEVNULL,

"curl", "-sL", "--max-time", "30",                     "-o", pdf_path, f"https://arxiv.org/pdf/{arxiv_id}.pdf",                     stdout=asyncio.subprocess.DEVNULL,

"-o", pdf_path, f"https://arxiv.org/pdf/{arxiv_id}.pdf",

"-o", pdf_path, f"https://arxiv.org/pdf/{arxiv_id}.pdf",

#!/usr/bin/env python3

#!/usr/bin/env python3

Architecture:     - asyncio + subprocess curl for concurrent HTTP requests

Architecture:     - asyncio + subprocess curl for concurrent HTTP requests

- asyncio + subprocess curl for concurrent HTTP requests

- asyncio + subprocess curl for concurrent HTTP requests

- Semaphore(10) to avoid hammering arXiv     - Pure regex HTML parsing (no WebFetch / no external deps)     - Per-request timeout via curl --max-time (no Python-level per-paper timeout)

- Semaphore(10) to avoid hammering arXiv     - Pure regex HTML parsing (no WebFetch / no external deps)     - Per-request timeout via curl --max-time (no Python-level per-paper timeout)

- Pure regex HTML parsing (no WebFetch / no external deps)     - Per-request timeout via curl --max-time (no Python-level per-paper timeout) """

- Pure regex HTML parsing (no WebFetch / no external deps)     - Per-request timeout via curl --max-time (no Python-level per-paper timeout) """

- Per-request timeout via curl --max-time (no Python-level per-paper timeout)

- Per-request timeout via curl --max-time (no Python-level per-paper timeout)

"""Fetch URL content using curl subprocess with retry. Returns empty string on failure."""

"""Fetch URL content using curl subprocess with retry. Returns empty string on failure."""

url = fig         if url.startswith("/"):             url = "https://arxiv.org" + url

url = fig         if url.startswith("/"):             url = "https://arxiv.org" + url

if url.startswith("/"):             url = "https://arxiv.org" + url         elif not url.startswith("http"):

if url.startswith("/"):             url = "https://arxiv.org" + url         elif not url.startswith("http"):

url = "https://arxiv.org" + url

url = "https://arxiv.org" + url

elif not url.startswith("http"):             if re.match(r"\d{4}\.\d{4,5}v\d+/", url):                 url = "https://arxiv.org/html/" + url

elif not url.startswith("http"):             if re.match(r"\d{4}\.\d{4,5}v\d+/", url):                 url = "https://arxiv.org/html/" + url

if re.match(r"\d{4}\.\d{4,5}v\d+/", url):                 url = "https://arxiv.org/html/" + url             else:

if re.match(r"\d{4}\.\d{4,5}v\d+/", url):                 url = "https://arxiv.org/html/" + url             else:

url = "https://arxiv.org/html/" + url

url = "https://arxiv.org/html/" + url

else:                 url = f"https://arxiv.org/html/{arxiv_id}/" + url         return url

else:                 url = f"https://arxiv.org/html/{arxiv_id}/" + url         return url

url = f"https://arxiv.org/html/{arxiv_id}/" + url

url = f"https://arxiv.org/html/{arxiv_id}/" + url

try:                 cmd = (                     f'curl -sL --max-time {CURL_TIMEOUT} "https://arxiv.org/pdf/{arxiv_id}"'

try:                 cmd = (                     f'curl -sL --max-time {CURL_TIMEOUT} "https://arxiv.org/pdf/{arxiv_id}"'

cmd = (                     f'curl -sL --max-time {CURL_TIMEOUT} "https://arxiv.org/pdf/{arxiv_id}"'                     f" | pdftotext -l 2 - -"

cmd = (                     f'curl -sL --max-time {CURL_TIMEOUT} "https://arxiv.org/pdf/{arxiv_id}"'                     f" | pdftotext -l 2 - -"

f'curl -sL --max-time {CURL_TIMEOUT} "https://arxiv.org/pdf/{arxiv_id}"'

f'curl -sL --max-time {CURL_TIMEOUT} "https://arxiv.org/pdf/{arxiv_id}"'

html_url = f"https://arxiv.org/html/{arxiv_id}"

html_url = f"https://arxiv.org/html/{arxiv_id}"

abs_affiliations = []         if not html_authors or not html_affiliations:             abs_url = f"https://arxiv.org/abs/{arxiv_id}"

abs_affiliations = []         if not html_authors or not html_affiliations:             abs_url = f"https://arxiv.org/abs/{arxiv_id}"

if not html_authors or not html_affiliations:             abs_url = f"https://arxiv.org/abs/{arxiv_id}"             abs_html = await curl_fetch(abs_url, sem)

if not html_authors or not html_affiliations:             abs_url = f"https://arxiv.org/abs/{arxiv_id}"             abs_html = await curl_fetch(abs_url, sem)

abs_url = f"https://arxiv.org/abs/{arxiv_id}"

abs_url = f"https://arxiv.org/abs/{arxiv_id}"

# figure_url: HTML curl > keep existing

# figure_url: HTML curl > keep existing

#!/usr/bin/env python3

Usage:     curl -sL "https://arxiv.org/pdf/{arxiv_id}" | pdftotext -l 2 - - | python3 extract_affiliations.py

curl -sL "https://arxiv.org/pdf/{arxiv_id}" | pdftotext -l 2 - - | python3 extract_affiliations.py

#!/usr/bin/env python3

#!/usr/bin/env python3

Executable file; content unavailable

ATOM_NS = {     "atom": "http://www.w3.org/2005/Atom",

ATOM_NS = {     "atom": "http://www.w3.org/2005/Atom",

"atom": "http://www.w3.org/2005/Atom",

"atom": "http://www.w3.org/2005/Atom",

"arxiv": "http://arxiv.org/schemas/atom",

"arxiv": "http://arxiv.org/schemas/atom",

"affiliations": "",         "abstract": p.get("summary", ""),         "url": f"https://arxiv.org/abs/{arxiv_id}",

"affiliations": "",         "abstract": p.get("summary", ""),         "url": f"https://arxiv.org/abs/{arxiv_id}",

"abstract": p.get("summary", ""),         "url": f"https://arxiv.org/abs/{arxiv_id}",         "pdf": f"https://arxiv.org/pdf/{arxiv_id}",

"abstract": p.get("summary", ""),         "url": f"https://arxiv.org/abs/{arxiv_id}",         "pdf": f"https://arxiv.org/pdf/{arxiv_id}",

"url": f"https://arxiv.org/abs/{arxiv_id}",

"url": f"https://arxiv.org/abs/{arxiv_id}",

"pdf": f"https://arxiv.org/pdf/{arxiv_id}",

"pdf": f"https://arxiv.org/pdf/{arxiv_id}",

while d <= end_date:             date_str = d.isoformat()             endpoint = f"https://huggingface.co/api/daily_papers?date={date_str}&limit=100"

while d <= end_date:             date_str = d.isoformat()             endpoint = f"https://huggingface.co/api/daily_papers?date={date_str}&limit=100"

date_str = d.isoformat()             endpoint = f"https://huggingface.co/api/daily_papers?date={date_str}&limit=100"             print(f"  Fetching hf-daily {date_str}...", file=sys.stderr)

date_str = d.isoformat()             endpoint = f"https://huggingface.co/api/daily_papers?date={date_str}&limit=100"             print(f"  Fetching hf-daily {date_str}...", file=sys.stderr)

endpoint = f"https://huggingface.co/api/daily_papers?date={date_str}&limit=100"

endpoint = f"https://huggingface.co/api/daily_papers?date={date_str}&limit=100"

endpoint = "https://huggingface.co/api/daily_papers?limit=50"

endpoint = "https://huggingface.co/api/daily_papers?limit=50"

endpoint = "https://huggingface.co/api/daily_papers?sort=trending&limit=50"

endpoint = "https://huggingface.co/api/daily_papers?sort=trending&limit=50"

cats = "+OR+".join(f"cat:{c}" for c in ARXIV_CATEGORIES)     url = (         f"https://export.arxiv.org/api/query?"

cats = "+OR+".join(f"cat:{c}" for c in ARXIV_CATEGORIES)     url = (         f"https://export.arxiv.org/api/query?"

url = (         f"https://export.arxiv.org/api/query?"         f"search_query=({cats})"

url = (         f"https://export.arxiv.org/api/query?"         f"search_query=({cats})"

f"https://export.arxiv.org/api/query?"

f"https://export.arxiv.org/api/query?"

"abstract": abstract,             "url": entry_url,             "pdf": f"https://arxiv.org/pdf/{arxiv_id}" if arxiv_id else "",

"abstract": abstract,             "url": entry_url,             "pdf": f"https://arxiv.org/pdf/{arxiv_id}" if arxiv_id else "",

"url": entry_url,             "pdf": f"https://arxiv.org/pdf/{arxiv_id}" if arxiv_id else "",             "date": date,

"url": entry_url,             "pdf": f"https://arxiv.org/pdf/{arxiv_id}" if arxiv_id else "",             "date": date,

"pdf": f"https://arxiv.org/pdf/{arxiv_id}" if arxiv_id else "",

"pdf": f"https://arxiv.org/pdf/{arxiv_id}" if arxiv_id else "",

parser = argparse.ArgumentParser()     parser.add_argument("--date", help="Target date YYYY-MM-DD (default: today)")     parser.add_argument("--days", type=int, default=1, help="Number of days to fetch (default: 1)")

parser = argparse.ArgumentParser()     parser.add_argument("--date", help="Target date YYYY-MM-DD (default: today)")     parser.add_argument("--days", type=int, default=1, help="Number of days to fetch (default: 1)")

parser.add_argument("--date", help="Target date YYYY-MM-DD (default: today)")     parser.add_argument("--days", type=int, default=1, help="Number of days to fetch (default: 1)")     parser.add_argument("--output", type=Path, help="Write UTF

parser.add_argument("--date", help="Target date YYYY-MM-DD (default: today)")     parser.add_argument("--days", type=int, default=1, help="Number of days to fetch (default: 1)")     parser.add_argument("--output", type=Path, help="Write UTF

parser.add_argument("--days", type=int, default=1, help="Number of days to fetch (default: 1)")

parser.add_argument("--days", type=int, default=1, help="Number of days to fetch (default: 1)")

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

[Credential-related source omitted]

#!/usr/bin/env python3

用法:     curl -s "https://export.arxiv.org/api/query?..." | python3 parse_arxiv.py

curl -s "https://export.arxiv.org/api/query?..." | python3 parse_arxiv.py

curl -s "https://export.arxiv.org/api/query?..." | python3 parse_arxiv.py

NS = {     "atom": "http://www.w3.org/2005/Atom",

"atom": "http://www.w3.org/2005/Atom",

"arxiv": "http://arxiv.org/schemas/atom",

"abstract": abstract,             "url": url,             "pdf": f"https://arxiv.org/pdf/{arxiv_id}" if arxiv_id else "",

"url": url,             "pdf": f"https://arxiv.org/pdf/{arxiv_id}" if arxiv_id else "",             "date": date,

"pdf": f"https://arxiv.org/pdf/{arxiv_id}" if arxiv_id else "",

#!/usr/bin/env python3

[Credential-related source omitted]

[Credential-related source omitted]

headers["Authorization"] = f"Bearer {token}"     url = "https://api.github.com/search/repositories?" + urlencode(         {"q": query, "sort": "stars", "order": "desc", "per_page": 10})

url = "https://api.github.com/search/repositories?" + urlencode(

#!/usr/bin/env python3

"owner": owner,                 "name": name,                 "url": f"https://github.com/{full}",

"name": name,                 "url": f"https://github.com/{full}",                 "description": description,

"url": f"https://github.com/{full}",

url = f"https://github.com/trending?since={args.since}"

#!/usr/bin/env python3

1. Determine the correct arXiv ID and source figure count. 2. Prefer `https://arxiv.org/html/ID`: extract figure captions and actual image URLs.

2. Prefer `https://arxiv.org/html/ID`: extract figure captions and actual image URLs.

|----------|------|----------| | PDF 路径 | `/path/to/paper.pdf` | 直接 Read | | arXiv 链接（可单独发送） | `https://arxiv.org/abs/xxxx`、`/pdf/xxxx`、`/html/xxxx` | 解析 ID/版本 → 优先获取 HTML 全文，失败回退 PDF → 完整阅读并保存 |

|----------|------|----------| | PDF 路径 | `/path/to/paper.pdf` | 直接 Read | | arXiv 链接（可单独发送） | `https://arxiv.org/abs/xxxx`、`/pdf/xxxx`、`/html/xxxx` | 解析 ID/版本 → 优先获取 HTML 全文，失败回退 PDF → 完整阅读并保存 |

| PDF 路径 | `/path/to/paper.pdf` | 直接 Read | | arXiv 链接（可单独发送） | `https://arxiv.org/abs/xxxx`、`/pdf/xxxx`、`/html/xxxx` | 解析 ID/版本 → 优先获取 HTML 全文，失败回退 PDF → 完整阅读并保存 | | Zotero 分类 | "VLA 分类的论文" | 查询数据库 → 列出 → 用户选择 |

| PDF 路径 | `/path/to/paper.pdf` | 直接 Read | | arXiv 链接（可单独发送） | `https://arxiv.org/abs/xxxx`、`/pdf/xxxx`、`/html/xxxx` | 解析 ID/版本 → 优先获取 HTML 全文，失败回退 PDF → 完整阅读并保存 | | Zotero 分类 | "VLA 分类的论文" | 查询数据库 → 列出 → 用户选择 |

| arXiv 链接（可单独发送） | `https://arxiv.org/abs/xxxx`、`/pdf/xxxx`、`/html/xxxx` | 解析 ID/版本 → 优先获取 HTML 全文，失败回退 PDF → 完整阅读并保存 |

| arXiv 链接（可单独发送） | `https://arxiv.org/abs/xxxx`、`/pdf/xxxx`、`/html/xxxx` | 解析 ID/版本 → 优先获取 HTML 全文，失败回退 PDF → 完整阅读并保存 |

2. 按优先级获取：arXiv HTML > arXiv PDF > DOI > WebSearch 标题 3. 判断 arXiv ID：从 URL / Zotero extra 字段 / 标题搜索 4. 推荐直接 WebFetch `https://arxiv.org/html/{arxiv_id}`，无需下载

2. 按优先级获取：arXiv HTML > arXiv PDF > DOI > WebSearch 标题 3. 判断 arXiv ID：从 URL / Zotero extra 字段 / 标题搜索 4. 推荐直接 WebFetch `https://arxiv.org/html/{arxiv_id}`，无需下载

3. 判断 arXiv ID：从 URL / Zotero extra 字段 / 标题搜索 4. 推荐直接 WebFetch `https://arxiv.org/html/{arxiv_id}`，无需下载 5. 跳过条件：既无 PDF 也无在线来源 / 非论文内容

3. 判断 arXiv ID：从 URL / Zotero extra 字段 / 标题搜索 4. 推荐直接 WebFetch `https://arxiv.org/html/{arxiv_id}`，无需下载 5. 跳过条件：既无 PDF 也无在线来源 / 非论文内容

4. 推荐直接 WebFetch `https://arxiv.org/html/{arxiv_id}`，无需下载

4. 推荐直接 WebFetch `https://arxiv.org/html/{arxiv_id}`，无需下载

1. WebSearch `"{论文标题} arxiv"` 获取 arXiv ID 2. **来源 A — arXiv HTML**（首选）：    - WebFetch `https://arxiv.org/html/{arxiv_id}` 提取所有 `<figure>` 的标题与 img src URL

1. WebSearch `"{论文标题} arxiv"` 获取 arXiv ID 2. **来源 A — arXiv HTML**（首选）：    - WebFetch `https://arxiv.org/html/{arxiv_id}` 提取所有 `<figure>` 的标题与 img src URL

2. **来源 A — arXiv HTML**（首选）：    - WebFetch `https://arxiv.org/html/{arxiv_id}` 提取所有 `<figure>` 的标题与 img src URL    - 统计论文 Figure 总数，确认提取数量是否完整

2. **来源 A — arXiv HTML**（首选）：    - WebFetch `https://arxiv.org/html/{arxiv_id}` 提取所有 `<figure>` 的标题与 img src URL    - 统计论文 Figure 总数，确认提取数量是否完整

- WebFetch `https://arxiv.org/html/{arxiv_id}` 提取所有 `<figure>` 的标题与 img src URL

- WebFetch `https://arxiv.org/html/{arxiv_id}` 提取所有 `<figure>` 的标题与 img src URL

5. 笔记中用 `![Figure X](url)` 外链嵌入 6. 验证：外链可加载 / 本地文件 >10KB 7. **URL 去重（必做）**：写入前检查每个 URL，如果 `https://arxiv.org/html/` 后出现了连续两段相同的 arxiv_id（如 `2607.04988v1/2607.04988v1/`），删除重复段。这是高频 bug，详见 `references/image-troubleshooting.md`

5. 笔记中用 `![Figure X](url)` 外链嵌入 6. 验证：外链可加载 / 本地文件 >10KB 7. **URL 去重（必做）**：写入前检查每个 URL，如果 `https://arxiv.org/html/` 后出现了连续两段相同的 arxiv_id（如 `2607.04988v1/2607.04988v1/`），删除重复段。这是高频 bug，详见 `references/image-troubleshooting.md`

6. 验证：外链可加载 / 本地文件 >10KB 7. **URL 去重（必做）**：写入前检查每个 URL，如果 `https://arxiv.org/html/` 后出现了连续两段相同的 arxiv_id（如 `2607.04988v1/2607.04988v1/`），删除重复段。这是高频 bug，详见 `references/image-troubleshooting.md`    - 原因：arXiv HTML 中 img src 已含 `{id}v1/` 前缀，拼接

6. 验证：外链可加载 / 本地文件 >10KB 7. **URL 去重（必做）**：写入前检查每个 URL，如果 `https://arxiv.org/html/` 后出现了连续两段相同的 arxiv_id（如 `2607.04988v1/2607.04988v1/`），删除重复段。这是高频 bug，详见 `references/image-troubleshooting.md`    - 原因：arXiv HTML 中 img src 已含 `{id}v1/` 前缀，拼接

7. **URL 去重（必做）**：写入前检查每个 URL，如果 `https://arxiv.org/html/` 后出现了连续两段相同的 arxiv_id（如 `2607.04988v1/2607.04988v1/`），删除重复段。这是高频 bug，详见 `references/image-troubleshooting.md`

7. **URL 去重（必做）**：写入前检查每个 URL，如果 `https://arxiv.org/html/` 后出现了连续两段相同的 arxiv_id（如 `2607.04988v1/2607.04988v1/`），删除重复段。这是高频 bug，详见 `references/image-troubleshooting.md`

- 原因：arXiv HTML 中 img src 已含 `{id}v1/` 前缀，拼接时不要再加一次    - 正确：`https://arxiv.org/html/` + src（如 `https://arxiv.org/html/2607.04988v1/x1.png`）    - 错误：`https://arxiv.org/html/2607.04988v1/` + src（导致路径重复，返回 404）

- 原因：arXiv HTML 中 img src 已含 `{id}v1/` 前缀，拼接时不要再加一次    - 正确：`https://arxiv.org/html/` + src（如 `https://arxiv.org/html/2607.04988v1/x1.png`）    - 错误：`https://arxiv.org/html/2607.04988v1/` + src（导致路径重复，返回 404）

- 正确：`https://arxiv.org/html/` + src（如 `https://arxiv.org/html/2607.04988v1/x1.png`）

- 正确：`https://arxiv.org/html/` + src（如 `https://arxiv.org/html/2607.04988v1/x1.png`）

- 错误：`https://arxiv.org/html/2607.04988v1/` + src（导致路径重复，返回 404）

- 错误：`https://arxiv.org/html/2607.04988v1/` + src（导致路径重复，返回 404）

#!/usr/bin/env python3

#!/usr/bin/env python3

#!/usr/bin/env python3

[Credential-related source omitted]

if arxiv_id:         source_lines.append(f"arXiv ID: {arxiv_id}")         source_lines.append(f"arXiv 页面: https://arxiv.org/abs/{arxiv_id}")

source_lines.append(f"arXiv ID: {arxiv_id}")         source_lines.append(f"arXiv 页面: https://arxiv.org/abs/{arxiv_id}")         source_lines.append(f"arXiv PDF: https://arxiv.org/pdf/{arxiv_id}.pdf")

source_lines.append(f"arXiv 页面: https://arxiv.org/abs/{arxiv_id}")

source_lines.append(f"arXiv PDF: https://arxiv.org/pdf/{arxiv_id}.pdf")

source_lines.append(f"arXiv HTML (图片): https://arxiv.org/html/{arxiv_id}")

if paper_source.get('doi'):         source_lines.append(f"DOI: {paper_source['doi']}")         source_lines.append(f"DOI 链接: https://doi.org/{paper_source['doi']}")

source_lines.append(f"DOI: {paper_source['doi']}")         source_lines.append(f"DOI 链接: https://doi.org/{paper_source['doi']}")     if paper_source.get('url'):

source_lines.append(f"DOI 链接: https://doi.org/{paper_source['doi']}")

fallback_steps.extend(                 [                     f"1. **arXiv HTML 版本**（推荐）: 用 WebFetch 读取 https://arxiv.org/html/{arxiv_id}，可直接获取图片 URL",

[                     f"1. **arXiv HTML 版本**（推荐）: 用 WebFetch 读取 https://arxiv.org/html/{arxiv_id}，可直接获取图片 URL",                     f"2. **arXiv 摘要页**: 用 WebFetch 读取 https://arxiv.org/abs/{arxiv_id}",

f"1. **arXiv HTML 版本**（推荐）: 用 WebFetch 读取 https://arxiv.org/html/{arxiv_id}，可直接获取图片 URL",

f"2. **arXiv 摘要页**: 用 WebFetch 读取 https://arxiv.org/abs/{arxiv_id}",

f"3. **arXiv PDF**: 下载 https://arxiv.org/pdf/{arxiv_id}.pdf 到本地后用 Read 读取",

)         if paper_source.get('doi'):             fallback_steps.append(f"{len(fallback_steps) + 1}. **DOI 页面**: 跳转到 https://doi.org/{paper_source['doi']} 读取")

if paper_source.get('doi'):             fallback_steps.append(f"{len(fallback_steps) + 1}. **DOI 页面**: 跳转到 https://doi.org/{paper_source['doi']} 读取")         if paper_source.get('url'):

fallback_steps.append(f"{len(fallback_steps) + 1}. **DOI 页面**: 跳转到 https://doi.org/{paper_source['doi']} 读取")

try:         result = subprocess.run(             ['claude', '-p', prompt, '--model', 'opus', '--permission-mode', 'acceptEdits', '--dangerously-skip-permissions'],

result = subprocess.run(             ['claude', '-p', prompt, '--model', 'opus', '--permission-mode', 'acceptEdits', '--dangerously-skip-permissions'],             capture_output=True,

['claude', '-p', prompt, '--model', 'opus', '--permission-mode', 'acceptEdits', '--dangerously-skip-permissions'],

### 来源 A: arXiv HTML（首选） - WebFetch `https://arxiv.org/html/{arxiv_id}` → 提取 `<figure>` 的 img src

- WebFetch `https://arxiv.org/html/{arxiv_id}` → 提取 `<figure>` 的 img src

curl -L -o "{TEMP_DIR}/paper.pdf" "https://arxiv.org/pdf/{arxiv_id}.pdf"

| src 值 | 正确拼接 | 错误拼接 | |--------|----------|----------| | `2607.04988v1/x1.png` | `https://arxiv.org/html/` + src | `https://arxiv.org/html/2607.04988v1/` + src |

|--------|----------|----------| | `2607.04988v1/x1.png` | `https://arxiv.org/html/` + src | `https://arxiv.org/html/2607.04988v1/` + src | | `figures/fig1.png` | `https://arxiv.org/html/2607.04988v1/` + src | — |

| `2607.04988v1/x1.png` | `https://arxiv.org/html/` + src | `https://arxiv.org/html/2607.04988v1/` + src |

| `figures/fig1.png` | `https://arxiv.org/html/2607.04988v1/` + src | — |

**铁律**: 1. 如果 src 以 `2` 开头（含 arxiv_id 前缀），base = `https://arxiv.org/html/`

1. 如果 src 以 `2` 开头（含 arxiv_id 前缀），base = `https://arxiv.org/html/`

2. 如果 src 不含 arxiv_id 前缀（如 `figures/`），base = `https://arxiv.org/html/{arxiv_id}v1/`

✗ https://arxiv.org/html/2607.04988v1/2607.04988v1/x1.png  ← 路径重复，404

✓ https://arxiv.org/html/2607.04988v1/x1.png               ← 正确

![Figure 1](https://arxiv.org/html/xxxx/x1.png)