deepxiv

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

DeepXiv Paper Search & Progressive Reading

DeepXiv 论文搜索与渐进式阅读

Search topic or paper ID: $ARGUMENTS

搜索主题或论文ID：$ARGUMENTS

Role & Positioning

角色与定位

DeepXiv is the progressive-reading literature source:

Skill	Best for
`/arxiv`	Direct preprint search and PDF download
`/semantic-scholar`	Published venue metadata, citation counts, DOI links
`/deepxiv`	Layered reading: search → brief → head → section, plus trending and web search

Use DeepXiv when you want to avoid loading full papers too early.

DeepXiv是一款支持渐进式阅读的文献资源工具：

技能	适用场景
`/arxiv`	直接预印本搜索与PDF下载
`/semantic-scholar`	获取已发表期刊元数据、引用次数、DOI链接
`/deepxiv`	分层阅读：搜索→摘要→概览→章节，外加热门论文检索与网页搜索

当你想避免过早加载完整论文时，请使用DeepXiv。

Constants

常量定义

FETCH_SCRIPT —
```
tools/deepxiv_fetch.py
```
relative to the current project. If unavailable, fall back to the raw
```
deepxiv
```
CLI.
MAX_RESULTS = 10 — Default number of results to return.

Overrides (append to arguments):
/deepxiv "agent memory" - max: 5
— top 5 results
/deepxiv "2409.05591" - brief
— quick paper summary
/deepxiv "2409.05591" - head
— metadata + section overview
/deepxiv "2409.05591" - section: Introduction
— read one section only
/deepxiv "trending" - days: 14 - max: 10
— trending papers
/deepxiv "karpathy" - web
— DeepXiv web search
/deepxiv "258001" - sc
— Semantic Scholar metadata by ID

FETCH_SCRIPT — 当前项目下的
```
tools/deepxiv_fetch.py
```
。若不可用，则退回到原生
```
deepxiv
```
CLI工具。
MAX_RESULTS = 10 — 默认返回的结果数量。

参数覆盖（追加到参数后）：
/deepxiv "agent memory" - max: 5
— 返回前5条结果
/deepxiv "2409.05591" - brief
— 快速获取论文摘要
/deepxiv "2409.05591" - head
— 获取元数据+章节概览
/deepxiv "2409.05591" - section: Introduction
— 仅阅读指定章节
/deepxiv "trending" - days: 14 - max: 10
— 获取热门论文
/deepxiv "karpathy" - web
— 使用DeepXiv进行网页搜索
/deepxiv "258001" - sc
— 通过ID获取Semantic Scholar元数据

Setup

安装配置

DeepXiv is optional. If the CLI is not installed, tell the user:

bash

pip install deepxiv-sdk

On first use,

deepxiv

auto-registers a free token and stores it in

~/.env

DeepXiv为可选工具。若未安装CLI工具，请告知用户执行以下命令：

bash

pip install deepxiv-sdk

首次使用时，

deepxiv

会自动注册免费令牌并存储在

~/.env

文件中。

Workflow

工作流程

Step 1: Parse Arguments

步骤1：解析参数

Parse

$ARGUMENTS

for:

Query or ID: a paper topic, arXiv ID, or Semantic Scholar ID
- max: N
: override
```
MAX_RESULTS
```
- brief
: fetch paper brief
- head
: fetch metadata and section map
- section: NAME
: fetch one named section
- trending
or query
```
trending
```
: fetch trending papers
- days: 7|14|30
: trending time window
- web
: run DeepXiv web search
- sc
: fetch Semantic Scholar metadata by ID

If the main argument looks like an arXiv ID and no explicit mode is given, default to

- brief

解析

$ARGUMENTS

中的内容：

查询词或ID：论文主题、arXiv ID或Semantic Scholar ID
- max: N
：覆盖
```
MAX_RESULTS
```
的默认值
- brief
：获取论文摘要
- head
：获取元数据与章节结构
- section: NAME
：获取指定章节内容
- trending
或查询词为
```
trending
```
：获取热门论文
- days: 7|14|30
：热门论文的时间范围
- web
：执行DeepXiv网页搜索
- sc
：通过ID获取Semantic Scholar元数据

若主参数看起来是arXiv ID且未指定明确模式，默认使用

- brief

模式。

Step 2: Locate the Adapter

步骤2：定位适配脚本

Prefer the ARIS adapter:

bash

python3 tools/deepxiv_fetch.py --help

tools/deepxiv_fetch.py

is not available, fall back to raw

deepxiv

commands.

优先使用ARIS适配脚本：

bash

python3 tools/deepxiv_fetch.py --help

若

tools/deepxiv_fetch.py

不可用，则退回到原生

deepxiv

命令。

Step 3: Execute the Minimal Command

步骤3：执行最简命令

Search papers

bash

python3 tools/deepxiv_fetch.py search "QUERY" --max MAX_RESULTS

Fallback:

bash

deepxiv search "QUERY" --limit MAX_RESULTS --format json

Brief summary

bash

python3 tools/deepxiv_fetch.py paper-brief ARXIV_ID

Fallback:

bash

deepxiv paper ARXIV_ID --brief --format json

Section map

bash

python3 tools/deepxiv_fetch.py paper-head ARXIV_ID

Fallback:

bash

deepxiv paper ARXIV_ID --head --format json

Specific section

bash

python3 tools/deepxiv_fetch.py paper-section ARXIV_ID "SECTION_NAME"

Fallback:

bash

deepxiv paper ARXIV_ID --section "SECTION_NAME" --format json

Trending

bash

python3 tools/deepxiv_fetch.py trending --days 7 --max MAX_RESULTS

Fallback:

bash

deepxiv trending --days 7 --limit MAX_RESULTS --output json

Web search

bash

python3 tools/deepxiv_fetch.py wsearch "QUERY"

Fallback:

bash

deepxiv wsearch "QUERY" --output json

Semantic Scholar metadata

bash

python3 tools/deepxiv_fetch.py sc "SEMANTIC_SCHOLAR_ID"

Fallback:

bash

deepxiv sc "SEMANTIC_SCHOLAR_ID" --output json

搜索论文

bash

python3 tools/deepxiv_fetch.py search "QUERY" --max MAX_RESULTS

备选命令：

bash

deepxiv search "QUERY" --limit MAX_RESULTS --format json

获取论文摘要

bash

python3 tools/deepxiv_fetch.py paper-brief ARXIV_ID

备选命令：

bash

deepxiv paper ARXIV_ID --brief --format json

获取章节结构

bash

python3 tools/deepxiv_fetch.py paper-head ARXIV_ID

备选命令：

bash

deepxiv paper ARXIV_ID --head --format json

获取指定章节

bash

python3 tools/deepxiv_fetch.py paper-section ARXIV_ID "SECTION_NAME"

备选命令：

bash

deepxiv paper ARXIV_ID --section "SECTION_NAME" --format json

获取热门论文

bash

python3 tools/deepxiv_fetch.py trending --days 7 --max MAX_RESULTS

备选命令：

bash

deepxiv trending --days 7 --limit MAX_RESULTS --output json

网页搜索

bash

python3 tools/deepxiv_fetch.py wsearch "QUERY"

备选命令：

bash

deepxiv wsearch "QUERY" --output json

获取Semantic Scholar元数据

bash

python3 tools/deepxiv_fetch.py sc "SEMANTIC_SCHOLAR_ID"

备选命令：

bash

deepxiv sc "SEMANTIC_SCHOLAR_ID" --output json

Step 4: Present Results

步骤4：展示结果

When searching, present a compact table:

text

| # | ID | Title | Year | Citations | Notes |
|---|----|-------|------|-----------|-------|

When reading a paper, show:

title
arXiv ID
authors
venue/date if available
TLDR or abstract summary
suggested next step:
```
brief
```
→
```
head
```
→
```
section
```

搜索时，展示简洁表格：

text

| # | ID | 标题 | 年份 | 引用次数 | 备注 |
|---|----|-------|------|-----------|-------|

阅读论文时，展示以下内容：

标题
arXiv ID
作者
期刊/发表日期（若有）
TLDR或摘要总结
建议下一步操作：
```
brief
```
→
```
head
```
→
```
section
```

Step 5: Escalate Depth Only When Needed

步骤5：仅在必要时加深阅读深度

Use this progression:

```
search
```
```
paper-brief
```
```
paper-head
```
```
paper-section
```
full paper only if necessary

Do not jump to full-paper reads when a brief or one section answers the question.

遵循以下递进流程：

```
search
```
（搜索）
```
paper-brief
```
（论文摘要）
```
paper-head
```
（论文概览）
```
paper-section
```
（指定章节）
仅在必要时查看完整论文

当摘要或单个章节即可回答问题时，不要直接跳转到完整论文阅读。

Step 6: Update Research Wiki (if active)

步骤6：更新研究维基（若启用）

Required when
research-wiki/
exists in the project; skip silently otherwise. Ingest papers that were meaningfully read (brief / head / section / full) during this invocation — mere

search

hits without a depth read do not need ingestion:

if [ -d research-wiki/ ]:
    for each arxiv_id the user asked this skill to read in depth:
        python3 tools/research_wiki.py ingest_paper research-wiki/ \
            --arxiv-id "<arxiv_id>"

The helper handles metadata / slug / dedup / page / index / log in one call — do not handwrite
papers/<slug>.md
. See

shared-references/integration-contract.md

. Backfill missed ingests with

python3 tools/research_wiki.py sync research-wiki/ --arxiv-ids <id1>,<id2>,...

当项目中存在
research-wiki/
目录时必须执行；否则静默跳过。将本次调用中被深度阅读的论文（摘要/概览/章节/完整论文）纳入维基——仅搜索但未深度阅读的结果无需纳入：

if [ -d research-wiki/ ]:
    for each arxiv_id the user asked this skill to read in depth:
        python3 tools/research_wiki.py ingest_paper research-wiki/ \
            --arxiv-id "<arxiv_id>"

该辅助工具会一次性处理元数据/别名/去重/页面/索引/日志——请勿手动编写
papers/<slug>.md
文件。详情请参阅

shared-references/integration-contract.md

。若有遗漏的纳入操作，可使用以下命令补全：

python3 tools/research_wiki.py sync research-wiki/ --arxiv-ids <id1>,<id2>,...

Key Rules

核心规则

Prefer the adapter script over raw
```
deepxiv
```
commands when available.
DeepXiv is optional. If unavailable, give the install command and suggest
```
/arxiv
```
or
```
/research-lit "topic" - sources: web
```
.
Use section-level reads to save tokens.
Treat DeepXiv as complementary to
```
/arxiv
```
and
```
/semantic-scholar
```
, not a replacement.
If the result overlaps with a published venue paper from Semantic Scholar, keep the richer venue metadata in the final summary.

若适配脚本可用，优先使用而非原生
```
deepxiv
```
命令。
DeepXiv为可选工具。若不可用，提供安装命令并建议使用
```
/arxiv
```
或
```
/research-lit "topic" - sources: web
```
。
使用按章节阅读的方式节省令牌消耗。
将DeepXiv视为
```
/arxiv
```
和
```
/semantic-scholar
```
的补充工具，而非替代品。
若结果与Semantic Scholar中的已发表期刊论文重叠，请在最终总结中保留更丰富的期刊元数据。