acquire-research-papers

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Acquire Research Papers

获取研究论文

Deliver the official publisher PDF and the publisher's raw BibTeX export as a verified pair. Discovery metadata nominates candidates and supports comparison; it never becomes the final citation artifact.

交付经核验的官方出版商 PDF 与出版商原始 BibTeX 导出文件，二者必须成对。Discovery 元数据仅用于提名候选论文并支持比较；它绝不会成为最终引用成果。

Run the installed skill

运行已安装的 Skill

Use this shared skill in Codex, Claude Code, Kiro IDE, or Kiro CLI. Resolve its actual installed directory, then run its portable launcher through the selected host's shell tool:

text

python <skill-directory>/scripts/arp.py --help
python <skill-directory>/scripts/arp.py doctor --host <actual-host> --skill-directory <skill-directory>
python <skill-directory>/scripts/arp.py config status

Python 3.11+ and

uv

are required. The standard-library launcher installs the fixed release wheel from

runtime.json

only after verifying SHA-256, in a separate cache outside the skill. No developer checkout or project environment is needed. A release wheel that has not yet been published cannot bootstrap from its release URL; a matching local release wheel may be supplied through

ARP_LOCAL_WHEEL

for isolated validation.

Read host installation for host paths, Kiro custom-agent resources, and checked support limits. The host doctor performs local checks by default; host loading, remote availability, and institutional access remain separate and unverified unless checked explicitly.

Check

arp config status

first. IEEE, MinerU, Elsevier, Semantic Scholar, and OpenAlex are optional and independent; configure only the capability needed for the current task. New installations use an editable local

.env

outside the repository:

arp config init

arp config path

, and

arp config edit

. A user may fill it locally or use non-echoing

config setup

prompts. Never ask for secret values in chat or command arguments. Legacy DPAPI installations remain legacy until explicit

arp config migrate --from dpapi

on the original Windows account. See credentials and cache.

Select the user's named institution profile for IEEE access. Stored configuration and historical checks do not prove current subscription or API entitlement. Public schema snapshots are included for reference;

arp schema <schema-name.json>

returns the authoritative schema bundled with the running release.

在 Codex、Claude Code、Kiro IDE 或 Kiro CLI 中使用此共享 Skill。解析其实际安装目录后，通过所选 host 的 shell 工具运行其便携启动器：

text

python <skill-directory>/scripts/arp.py --help
python <skill-directory>/scripts/arp.py doctor --host <actual-host> --skill-directory <skill-directory>
python <skill-directory>/scripts/arp.py config status

需要 Python 3.11+ 和

uv

。标准库启动器只有在验证 SHA-256 后，才会从

runtime.json

安装固定版本的 release wheel，并将其置于 Skill 目录之外的独立缓存中。无需开发者 checkout 或项目环境。尚未发布的 release wheel 无法从其发布 URL 引导安装；进行隔离验证时，可通过

ARP_LOCAL_WHEEL

提供匹配的本地 release wheel。

阅读 host 安装说明，了解 host 路径、Kiro custom-agent 资源以及已验证的支持限制。host doctor 默认执行本地检查；host 加载、远程可用性和机构访问权限彼此独立，除非明确检查，否则均未经验证。

首先检查

arp config status

。IEEE、MinerU、Elsevier、Semantic Scholar 和 OpenAlex 均为可选且相互独立；仅配置当前任务所需的能力。新安装使用仓库之外可编辑的本地

.env

：

arp config init

、

arp config path

和

arp config edit

。用户可以在本地填写，也可以使用不回显的

config setup

prompts。绝不在聊天或命令参数中索要密钥值。旧版 DPAPI 安装会保持 legacy 状态，直至在原 Windows 账户上显式执行

arp config migrate --from dpapi

。参阅 credentials 与 cache。

为 IEEE 访问选择用户指定的机构 profile。已存储的配置和历史检查不能证明当前订阅资格或 API 权限。文中附有公开 schema 快照供参考；

arp schema <schema-name.json>

会返回当前运行 release 中捆绑的权威 schema。

Route the task

任务路由

Request	Command and reference
Known DOI, publisher URL, exact title, or explicit paper list	`arp fetch` ; read acquisition workflows
Venue/topic/year/count corpus	`arp discover corpus --progress` → `arp review corpus` → `arp acquire corpus` ; read corpus mode
Nearest work, Related Work, claim citations, or gap evidence	`arp discover research` → review → acquire → evidence; read research mode

Start with copyable recipes. Related Work and gap analysis share one research workflow; choose the question type in the brief. Discovery writes state only. Evidence review produces

reports/report.md

; add

--export-csv

for tables. Never draft placeholder conclusions during discovery.

Advanced tasks: author bibliographies, library audit/update/import/organize and explicit chapters, independent capabilities and run recovery, institution profiles, and explicit Markdown export. Read only the reference needed for the task. Author and library audit default to JSON plus

report.md

; CSV is opt-in.

Before planning venue coverage, use

arp sources --spec <corpus.yaml>

and read sources and coverage. This offline diagnostic describes capability, not live coverage or universal completeness. Translate attached documents into the same explicit inputs. Keep corpus and research frozen hashes unchanged when running library or author tasks.

请求	命令与参考
已知 DOI、出版商 URL、准确标题或明确给出的论文列表	`arp fetch` ；阅读 acquisition workflows
按 venue/主题/年份/数量限定的 corpus	`arp discover corpus --progress` → `arp review corpus` → `arp acquire corpus` ；阅读 corpus mode
最相近的工作、Related Work、支撑主张的引用或研究空白证据	`arp discover research` → review → acquire → evidence；阅读 research mode

先从可复制 recipes 开始。Related Work 和空白分析共用同一 research 工作流；在任务简报中选择问题类型。Discovery 阶段仅写入状态。Evidence 审查会生成

reports/report.md

；如需表格，请添加

--export-csv

。Discovery 期间绝不起草占位结论。

高级任务包括：作者文献目录、论文库审查/更新/导入/整理与明确章节、独立能力与运行恢复、机构 profiles，以及显式 Markdown 导出。仅阅读当前任务所需的参考文档。Author 和 Library 审查默认输出 JSON 与

report.md

；CSV 为可选项。

在规划 venue coverage 之前，使用

arp sources --spec <corpus.yaml>

并阅读 sources 与 coverage。该离线诊断描述的是能力，而非实时覆盖情况或普遍完整性。将所附文档转换为相同的明确输入。运行 Library 或 Author 任务时，保持 corpus 和 research 的 frozen hash 不变。

Review and completion

审查与完成

Preserve original BibTeX bytes, BOM, encoding, and line endings. Decode only a parsing view. Never generate or repair citations with an LLM or a metadata API.
Verify DOI, title, year, author scope, and venue. Report ambiguous identities and partial evidence explicitly.
Corpus decisions inspect
```
evidence-packets.jsonl
```
and record title-and-abstract evidence. Acquire only the frozen
```
selected-papers.jsonl
```
bound by
```
selection-manifest.json
```
; keywords are optional and full text is not required for corpus relevance.
The executing agent makes semantic decisions and records the actual host. New v2 decisions set
```
reviewer
```
and
```
executor.host
```
to that host; include model/version only when actually available. Preserve legacy Codex v1 review bytes and hashes. Never relabel another host as Codex or invent model metadata. Identity, relevance, and full-text evidence policies remain distinct.
```
--defer-host
```
retains selections in
```
manual-download.csv
```
; never remove difficult papers from a frozen selection or lower relevance thresholds to fill a quota.
Library audits distinguish the local PDF version, citation placement, and current metadata. A current citation does not prove the PDF is current. Review
```
plan.json
```
before explicit
```
arp library apply --plan <path>
```
; update/import/organize do not apply automatically. Retain backups for rollback.
Author discovery reports source coverage, identity ambiguity, formal/early-access/preprint groups, and separate issue/online dates. Never claim an absolutely complete bibliography.
Ordinary fetch, manual fetch, and corpus workflows produce no Markdown. Research uses an internal MinerU cache; export Markdown only after an explicit user destination request.
A standalone
```
parse --pdf
```
needs no BibTeX but retains user-supplied trust. It does not satisfy the official-pair delivery contract. Capability
```
export --format markdown
```
formats structured records; it is separate from
```
export-md
```
of full text.
Capability runs record input/output hashes, item states, and next actions. Reuse the same
```
--run
```
and command inputs to resume; use
```
arp run status --run <directory>
```
to inspect them. A read-only adapter also summarizes supported legacy manifests without rewriting or converting them; legacy runs use their original resume commands. Agent review, user download, missing configuration, rate limits, retryable errors, and permanent failures remain distinct. Complete unrelated items without erasing blocked ones.

保留原始 BibTeX 字节、BOM、编码和行尾。仅为解析视图进行解码。绝不使用 LLM 或 metadata API 生成或修复引用。
核验 DOI、标题、年份、作者范围和 venue。明确报告身份歧义和不完整证据。
Corpus 决策需检查
```
evidence-packets.jsonl
```
，并记录标题与摘要证据。仅获取由
```
selection-manifest.json
```
绑定且已冻结的
```
selected-papers.jsonl
```
；关键词为可选项，corpus 相关性不要求全文。
执行任务的 Agent 作出语义决策并记录实际 host。新版 v2 决策将
```
reviewer
```
和
```
executor.host
```
设为该 host；只有在实际可用时才包含 model/version。保留 legacy Codex v1 review 的原始字节和 hash。绝不将其他 host 重新标记为 Codex，也绝不编造 model metadata。身份、相关性和全文证据策略保持彼此独立。
```
--defer-host
```
会将选择结果保留在
```
manual-download.csv
```
中；绝不从 frozen selection 中移除难以获取的论文，也绝不降低相关性阈值来凑足配额。
Library 审查区分本地 PDF 版本、引用位置和当前 metadata。引用是最新的，并不表示 PDF 也是当前版本。在显式执行
```
arp library apply --plan <path>
```
前审查
```
plan.json
```
；update/import/organize 不会自动应用。保留 backups 以便 rollback。
Author discovery 报告 source coverage、身份歧义、正式发表/early access/preprint 分组，以及彼此独立的 issue date 与 online date。绝不声称 bibliography 绝对完整。
常规 fetch、manual fetch 和 corpus 工作流均不生成 Markdown。Research 使用内部 MinerU cache；只有用户明确指定目标位置后才导出 Markdown。
独立运行的
```
parse --pdf
```
无需 BibTeX，但仍沿用用户提供的 trust。它不满足 official-pair 交付契约。Capability
```
export --format markdown
```
格式化结构化记录；它与全文
```
export-md
```
不同。
Capability run 会记录 input/output hash、item state 和 next action。恢复运行时，复用相同的
```
--run
```
和命令输入；使用
```
arp run status --run <directory>
```
检查状态。只读 adapter 仅汇总其支持的 legacy manifests，不会重写或转换这些 manifests；legacy run 使用原始 resume 命令。Agent review、user download、missing config、rate limit、retryable error 和 permanent failure 均保持为不同状态。完成其他未受阻条目时，不得擦除受阻条目。

Access boundaries

访问边界

Read source policies before changing an acquisition contract and configuration before institution adaptation.

Never print, log, commit, or send credentials, tokens, cookies, browser storage, or decrypted values. Keep state and deliveries outside the public repository. IEEE uses an isolated visible profile for the selected institution; credentials are released only at its exact configured hostname. A successful identity-provider login is separate from return to IEEE, a verified download, and proof of institutional entitlement.

Never automate ScienceDirect article pages or ScienceDirect organization-login flows, attach to the user's normal Chrome, read its profile, or export Cookie/session data. Authorized users download PDF and raw BibTeX themselves; the CLI handles local stable files afterward. Do not bypass publisher/institution controls or substitute mirrors, discovery BibTeX, or generated citations.

在更改 acquisition 契约之前阅读 source policies，并在适配机构访问之前阅读 configuration。

绝不打印、记录、提交或发送 credentials、tokens、cookies、browser storage 或解密后的值。将 state 和交付物保存在公共仓库之外。IEEE 为所选机构使用隔离的 visible profile；credentials 仅在配置中精确指定的 hostname 上释放。身份提供商登录成功，与成功返回 IEEE、经验证的下载及机构 entitlement 证明是相互独立的环节。

绝不自动操作 ScienceDirect 文章页面或 ScienceDirect 机构登录流程，绝不挂接用户日常使用的 Chrome、读取其 profile，或导出 Cookie/session 数据。授权用户自行下载 PDF 和原始 BibTeX；此后由 CLI 处理本地稳定文件。绝不绕过出版商或机构控制，也绝不用 mirrors、discovery BibTeX 或生成的 citations 进行替代。