asta-skill

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Asta MCP — Academic Paper Search

Asta MCP — 学术论文搜索

Asta is Ai2's Scientific Corpus Tool, exposing the Semantic Scholar academic graph over MCP (streamable HTTP transport). This skill tells agents which Asta tool to call for which intent, and how to compose them into useful workflows.

MCP endpoint:
```
https://asta-tools.allen.ai/mcp/v1
```
Auth:
```
x-api-key
```
header (request key at https://share.hsforms.com/1L4hUh20oT3mu8iXJQMV77w3ioxm)
Transport: streamable HTTP

Asta是Ai2推出的科学语料库工具，通过MCP（可流式传输的HTTP协议）对外暴露Semantic Scholar学术图谱。本技能指导Agent针对不同意图调用对应的Asta工具，以及如何将这些工具组合成实用的工作流。

MCP端点：
```
https://asta-tools.allen.ai/mcp/v1
```
认证方式：
```
x-api-key
```
请求头（可通过https://share.hsforms.com/1L4hUh20oT3mu8iXJQMV77w3ioxm申请密钥）
传输协议： 可流式传输的HTTP

Prerequisite Check

前置检查

Before invoking any tool, verify the Asta MCP server is registered in the host agent. Tool names will be prefixed by the MCP server name chosen at install time (commonly

asta__<tool>

mcp__asta__<tool>

). If no Asta tools are visible, direct the user to the Installation section below.

调用任何工具前，请确认Asta MCP服务器已在宿主Agent中注册。工具名称会带有安装时选择的MCP服务器名称前缀（通常为

asta__<tool>

或

mcp__asta__<tool>

）。如果未显示任何Asta工具，请引导用户查看下方的安装部分。

Tool Map — Intent → Asta Tool

工具映射 — 意图 → Asta工具

User intent	Asta tool	Notes
Broad topic search	`search_papers_by_relevance`	Supports venue + date filters
Known paper title	`search_paper_by_title`	Optional venue restriction
Known DOI / arXiv / PMID / CorpusId / MAG / ACL / SHA / URL	`get_paper`	Single-paper lookup
Multiple known IDs at once	`get_paper_batch`	Batch lookup — prefer over N sequential `get_paper` calls
Who cited paper X	`get_citations`	Citation traversal with filters, paginated
Find author by name	`search_authors_by_name`	Returns profile info
An author's publications	`get_author_papers`	Pass author id from previous call
Find passages mentioning X	`snippet_search`	~500-word excerpts from paper bodies

Search/citation tools accept publication_date_range
(format

YYYY-MM-DD:YYYY-MM-DD

; year shorthand like

"2021:"

":2015-01"

"2015:2020"

is also accepted) and venues
(comma-separated) filters, plus fields
for field selection — pass them whenever the user's intent constrains scope (e.g., "recent", "since 2022", "at NeurIPS").

用户意图	Asta工具	说明
宽泛主题搜索	`search_papers_by_relevance`	支持会议/期刊筛选 + 日期筛选
已知论文标题	`search_paper_by_title`	可选会议/期刊限制
已知DOI / arXiv / PMID / CorpusId / MAG / ACL / SHA / URL	`get_paper`	单篇论文查询
批量查询已知ID	`get_paper_batch`	批量查询 — 优先使用该工具，而非连续调用N次 `get_paper`
哪些论文引用了论文X	`get_citations`	带筛选条件的引用遍历，支持分页
通过姓名查找作者	`search_authors_by_name`	返回作者个人资料信息
作者的所有出版物	`get_author_papers`	传入上一步调用返回的作者ID
查找提及X的段落	`snippet_search`	论文正文中约500词的摘录片段

搜索/引用类工具支持**

publication_date_range

（格式为
YYYY-MM-DD:YYYY-MM-DD
；也支持年份简写，如
"2021:"
、
":2015-01"
、
"2015:2020"
）和

venues

（逗号分隔）筛选条件，以及用于字段选择的

fields

**参数——当用户意图存在范围限制时（例如“近期”、“2022年以来”、“NeurIPS会议”），请传入这些参数。

⚠️

fields

parameter — avoid context blowups

⚠️

fields

参数 — 避免上下文溢出

get_paper

get_paper_batch

accept a

fields

string. Never request
citations
or
references
via

fields

— a single highly-cited paper (e.g. Attention Is All You Need) returns 200k+ characters and will overflow the agent's context window. Use the dedicated

get_citations

tool for forward citations (it paginates). Asta does not provide a dedicated

get_references

tool — to retrieve a paper's reference list, use

get_paper

with

fields=references

only for papers you know have a small reference list (typically < 100).

Safe default

fields

for

get_paper

title,year,authors,venue,tldr,url,abstract

Add

journal

publicationDate

fieldsOfStudy

isOpenAccess

only when needed.

get_paper

get_paper_batch

支持

fields

字符串参数。切勿通过
fields
请求
citations
或
references
字段——一篇高引用论文（如《Attention Is All You Need》）会返回20万+字符，导致Agent的上下文窗口溢出。如需正向引用，请使用专门的

get_citations

工具（该工具支持分页）。Asta未提供专门的

get_references

工具——若要获取论文的参考文献列表，仅对已知参考文献数量较少的论文（通常少于100篇）使用

get_paper

并设置

fields=references

。

get_paper

的安全默认

fields

参数：

title,year,authors,venue,tldr,url,abstract

仅在需要时添加

journal

、

publicationDate

、

fieldsOfStudy

、

isOpenAccess

字段。

Retrieving DOI / external IDs (undocumented but supported)

获取DOI / 外部ID（未正式文档化但支持）

Asta's official

fields

list does not include

externalIds

, but the field is transparently passed through to the underlying Semantic Scholar API and works in practice. Add

externalIds

fields

to retrieve

DOI

PubMed

PubMedCentral

ArXiv

MAG

DBLP

CorpusId

. Caveats:

Not all papers have a DOI — pure arXiv preprints often only return
```
ArXiv
```
+
```
CorpusId
```
.
```
get_paper("DOI:...")
```
lookup is not 100% reliable; some valid DOIs return
```
not found
```
. Prefer searching by title first, then reading
```
externalIds
```
off the result.
Since this is undocumented, treat it as best-effort and degrade gracefully if a future Asta release drops it.

Asta官方的

fields

列表不包含
externalIds
，但该字段会透传给底层的Semantic Scholar API，实际可用。在

fields

中添加

externalIds

可获取

DOI

、

PubMed

、

PubMedCentral

、

ArXiv

、

MAG

、

DBLP

、

CorpusId

。注意事项：

并非所有论文都有DOI——纯arXiv预印本通常仅返回
```
ArXiv
```
+
```
CorpusId
```
。
通过
```
get_paper("DOI:...")
```
查询并非100%可靠；部分有效DOI会返回
```
not found
```
。建议先通过标题搜索，再从结果中读取
```
externalIds
```
。
由于该功能未正式文档化，请视为尽力而为的功能，若未来Asta版本移除该功能，请优雅降级处理。

Workflow Patterns

工作流模式

Pattern 1 — Topic Discovery

模式1 — 主题探索

```
search_papers_by_relevance(keyword, publication_date_range="<current_year-5>:", venues=?)
```
→ initial hits (compute the lower bound from today's date — e.g., in 2026 pass
```
publication_date_range="2021:"
```
; adjust or drop the filter if the user asks for older work)
Rank/present top N by citationCount + recency
Offer follow-ups:
```
get_citations
```
on the most influential, or
```
snippet_search
```
for specific claims

```
search_papers_by_relevance(keyword, publication_date_range="<当前年份-5>:", venues=?)
```
→ 初始搜索结果（根据当前日期计算下限——例如2026年传入
```
publication_date_range="2021:"
```
；若用户要求查找更早的文献，可调整或移除该筛选条件）
按引用量 + 时效性对前N篇结果排序/展示
提供后续操作选项：对最具影响力的论文执行
```
get_citations
```
，或针对特定观点执行
```
snippet_search
```

Pattern 2 — Seed-Paper Expansion

模式2 — 种子论文扩展

```
get_paper(DOI|arXiv|...)
```
→ verify seed
```
get_citations(paperId)
```
→ forward expansion
Optionally
```
search_papers_by_relevance
```
with seed title terms for sideways discovery
Deduplicate by paperId before presenting

```
get_paper(DOI|arXiv|...)
```
→ 验证种子论文
```
get_citations(paperId)
```
→ 正向扩展（查找引用该论文的文献）
可选：使用种子论文标题中的关键词执行
```
search_papers_by_relevance
```
进行横向探索
展示前先按paperId去重

Pattern 3 — Author Deep-Dive

模式3 — 作者深度挖掘

```
search_authors_by_name(name)
```
→ pick correct profile (disambiguate by affiliation)
```
get_author_papers(authorId)
```
→ full publication list
Filter client-side by topic keywords or date

```
search_authors_by_name(name)
```
→ 选择正确的作者资料（通过所属机构区分）
```
get_author_papers(authorId)
```
→ 获取完整的出版物列表
客户端侧按主题关键词或日期筛选结果

Pattern 4 — Evidence Retrieval

模式4 — 证据检索

```
snippet_search(claim_query)
```
→ find passages making/supporting a claim
For each hit, optionally
```
get_paper(id)
```
for full metadata

```
snippet_search(claim_query)
```
→ 查找提出/支持某观点的段落
对于每个结果，可选择执行
```
get_paper(id)
```
获取完整元数据

Output & Interaction Rules

输出与交互规则

Always report total count and which tool was used.
Present top 10 as a table (title, year, venue, citations), then details for the most relevant.
If the user writes in Chinese, present summaries in Chinese; keep titles in original language.
After results, offer: Details / Refine / Citations / Snippet / Export / Done.

始终报告结果总数和所使用的工具。
将前10条结果以表格形式展示（标题、年份、会议/期刊、引用量），然后展示最相关结果的详细信息。
如果用户使用中文提问，需用中文展示摘要；论文标题保留原语言。
展示结果后，提供选项：详情/筛选/引用/片段/导出/完成。

Critical Rules

关键规则

Prefer batched intent over ping-pong. If the user's question needs two independent lookups, issue them as parallel MCP tool calls in one turn, not sequentially.
Never guess IDs. If a user gives a fuzzy title, use
```
search_paper_by_title
```
before
```
get_paper
```
.
Respect rate limits. An API key buys higher limits but not unlimited — stop expanding citation graphs beyond what the user asked for.
Do not fabricate fields. If Asta returns null
```
abstract
```
or
```
venue
```
, say so rather than inventing.

优先批量处理，避免频繁交互。如果用户的问题需要两次独立查询，请在一轮中发起并行MCP工具调用，而非依次调用。
切勿猜测ID。如果用户提供模糊的标题，请先使用
```
search_paper_by_title
```
，再调用
```
get_paper
```
。
遵守速率限制。API密钥可提升速率限制，但并非无限制——不要超出用户需求扩展引用图谱。
切勿编造字段。如果Asta返回的
```
abstract
```
或
```
venue
```
为null，请如实告知，不要编造内容。

Handling Asta responses

处理Asta响应

Situation	What to do
Empty `abstract`	Not all corpus papers have full text — use `snippet_search` , or fall back to title + TLDR
Author disambiguation uncertain	Inspect affiliations in `search_authors_by_name` results before calling `get_author_papers`
`429 Too Many Requests`	Back off; batch with `get_paper_batch` instead of sequential `get_paper` calls
Need DOI / PubMed ID / arXiv ID	Add `externalIds` to `fields` (see "Retrieving DOI" above); fall back to `ArXiv` ID when `DOI` is absent

场景	处理方式
`abstract` 为空	并非所有语料库中的论文都有完整文本——可使用 `snippet_search` ，或退而展示标题 + TLDR
作者身份区分存疑	在调用 `get_author_papers` 前，检查 `search_authors_by_name` 结果中的所属机构信息
返回 `429 Too Many Requests`	暂停请求；改用 `get_paper_batch` 批量查询，而非连续调用 `get_paper`
需要DOI / PubMed ID / arXiv ID	在 `fields` 中添加 `externalIds` （见上方“获取DOI”部分）；若DOI不存在，退而使用 `ArXiv` ID

asta-skill

Original

Translation

Asta MCP — Academic Paper Search

Asta MCP — 学术论文搜索

Prerequisite Check

前置检查

Tool Map — Intent → Asta Tool

工具映射 — 意图 → Asta工具

⚠️
`fields`
parameter — avoid context blowups

⚠️
`fields`
参数 — 避免上下文溢出

Retrieving DOI / external IDs (undocumented but supported)

获取DOI / 外部ID（未正式文档化但支持）

Workflow Patterns

工作流模式

Pattern 1 — Topic Discovery

模式1 — 主题探索

Pattern 2 — Seed-Paper Expansion

模式2 — 种子论文扩展

Pattern 3 — Author Deep-Dive

模式3 — 作者深度挖掘

Pattern 4 — Evidence Retrieval

模式4 — 证据检索

Output & Interaction Rules

输出与交互规则

Critical Rules

关键规则

Handling Asta responses

处理Asta响应

asta-skill

Original

Translation

Asta MCP — Academic Paper Search

Asta MCP — 学术论文搜索

Prerequisite Check

前置检查

Tool Map — Intent → Asta Tool

工具映射 — 意图 → Asta工具

⚠️ fields parameter — avoid context blowups

⚠️ fields参数 — 避免上下文溢出

Retrieving DOI / external IDs (undocumented but supported)

获取DOI / 外部ID（未正式文档化但支持）

Workflow Patterns

工作流模式

Pattern 1 — Topic Discovery

模式1 — 主题探索

Pattern 2 — Seed-Paper Expansion

模式2 — 种子论文扩展

Pattern 3 — Author Deep-Dive

模式3 — 作者深度挖掘

Pattern 4 — Evidence Retrieval

模式4 — 证据检索

Output & Interaction Rules

输出与交互规则

Critical Rules

关键规则

Handling Asta responses

处理Asta响应

⚠️
`fields`
parameter — avoid context blowups

⚠️
`fields`
参数 — 避免上下文溢出