菜单
Fetch the FULL TEXT of an arXiv paper (all sections — introduction, method, results, conclusion). Use when you need to read beyond the abstract into the paper's actual content. Only works for arXiv papers. For metadata/abstract only, use paper-read. For local PDFs, use paper-read-pdf.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | paper-fetch |
| description | Fetch the FULL TEXT of an arXiv paper (all sections — introduction, method, results, conclusion). Use when you need to read beyond the abstract into the paper's actual content. Only works for arXiv papers. For metadata/abstract only, use paper-read. For local PDFs, use paper-read-pdf. |
| always | false |
Fetch the complete text of an arXiv paper using web_fetch via ar5iv.org (LaTeX → clean HTML),
with automatic fallback to the arXiv abstract page.
Primary: https://ar5iv.org/abs/<arxiv_id> — full rendered text (all sections)
Fallback: https://arxiv.org/abs/<arxiv_id> — abstract page only
ar5iv renders LaTeX source to HTML, giving clean readable text including equations, tables, and all sections. It is the best way to read arXiv papers as plain text.
| Input | Example |
|---|---|
| Bare arXiv ID | 1706.03762 |
| arXiv ID with prefix | arxiv:1706.03762 → strip prefix, use 1706.03762 |
| arXiv abstract URL | https://arxiv.org/abs/1706.03762 → extract ID 1706.03762 |
| arXiv PDF URL | https://arxiv.org/pdf/1706.03762 → extract ID 1706.03762 |
| ar5iv URL | https://ar5iv.org/abs/1706.03762 → use as-is |
Always normalize to bare ID before constructing URL.
From whatever format you receive, extract just the numeric ID:
arxiv:1706.03762 → 1706.03762https://arxiv.org/abs/2303.08774v2 → 2303.08774 (drop version suffix)ArXiv:2310.06825 → 2310.06825web_fetch("https://ar5iv.org/abs/1706.03762")
Good response: contains section headings like "Introduction", "Method", "Results" etc. Bad response / error: ar5iv page says "not available" or returns very little text.
If ar5iv returns insufficient content (< 500 chars of real text, or error):
web_fetch("https://arxiv.org/abs/1706.03762")
This gives title + abstract + metadata, but not full paper sections.
From a successful ar5iv response, extract and present:
If the user asks for a specific section, jump straight to it.
web_fetch("https://ar5iv.org/abs/1706.03762")
web_fetch("https://ar5iv.org/abs/2303.08774")
web_fetch("https://arxiv.org/abs/1706.03762")
openAccessPdf field obtained via paper-readpaper-read-pdf skill insteadIf you only need a specific section (e.g., just the experimental results), fetch the full ar5iv page and scan for the relevant heading. The text is structured with clear section titles.
Search HuggingFace Hub for datasets by keyword. Use when the user wants to find training data, benchmarks, or evaluation datasets for ML/NLP/CV research.
Two-layer memory system with grep-based recall for research sessions.
Track specific scientific claims across the literature over time — who made it, who replicated it, who challenged it, whether it still stands. Use when verifying a key assumption before building on it, or when checking whether a published result has been updated or superseded.
Scan papers for conflicting empirical claims, methodological disagreements, or opposing conclusions on the same topic. Use when writing discussion sections, evaluating conflicting results, or checking if a claim is contested before building on it.
Synthesize findings across multiple papers into a coherent narrative, structured comparison table, or temporal evolution. Use after collecting papers via survey or paper-search. Goes beyond summarizing individual papers to produce insights that only emerge when reading across the corpus as a whole.
Full end-to-end deep research pipeline on a topic. Use when the user wants thorough, rigorous research — not just a survey. Orchestrates all research skills in sequence: collect → synthesize → critique claims → grade evidence → find gaps → assess reproducibility → optionally reproduce → write report.