← Back to Skills Marketplace
文献采集
by
Tiandoufayale
· GitHub ↗
· v0.0.2
· MIT-0
224
Downloads
0
Stars
1
Active Installs
2
Versions
Install in OpenClaw
/install paper-collector
Description
按用户指定的研究方向检索并筛选学术论文：在期刊官网、预印本平台与学术来源中搜索，必须阅读摘要后才可纳入结果。适用于用户要求查找、收集、筛选或整理某一主题/研究问题/方向的文献；若未提供研究方向则不得执行。支持可选“仅限期刊清单”模式、可选论文数量覆盖、可选 Markdown 输出路径。
Usage Guidance
This skill appears to do what it claims, but before enabling it: 1) Verify whether the agent has web-browsing capability and what pages it may access (it will visit many publisher and index sites). 2) If you allow file writes, restrict or verify the provided .md output path so the agent cannot write to sensitive locations. 3) Be aware it may attempt to access paywalled pages (it only needs abstracts); decide how you want paywalled content handled. 4) Note the SKILL.md enforces a strict "recent one month" time filter (repeated in the file) — confirm this is intended. If you are comfortable with web access and controlled output paths, the skill is coherent and reasonable to use.
Capability Analysis
Type: OpenClaw Skill
Name: paper-collector
Version: 0.0.2
The 'paper-collector' skill is a legitimate research assistant tool designed to search, filter, and summarize academic papers based on user-defined topics. It utilizes standard web-browsing tools (e.g., agent-browser, tavily-search) to access reputable academic sources like PubMed and Nature, and it includes strict logic for validating research directions and reading abstracts before inclusion. No evidence of data exfiltration, malicious execution, or prompt injection was found in SKILL.md or _meta.json.
Capability Assessment
Purpose & Capability
Name/description match the instructions: it searches academic sources, reads abstracts, filters results, and optionally writes a Markdown file. It does not request unrelated credentials or binaries.
Instruction Scope
The SKILL.md instructs the agent to browse publisher sites, academic indexes, and preprint servers and to prefer certain third-party browsing/search helper skills if available. This is coherent for a paper-collector, but it grants the agent discretion to access many external web pages and to write a .md file to a user-provided path. Confirm the agent's browsing and filesystem permissions before use; the repeated rule limiting results to the 'last one month' is a strict retrieval filter that may be unintended but is not a security incoherence.
Install Mechanism
Instruction-only skill with no install spec and no code files. Nothing is downloaded or written during install.
Credentials
No environment variables, credentials, or config paths are requested — proportional to stated purpose.
Persistence & Privilege
always is false and the skill does not request persistent or cross-skill config changes. It may write an output .md when given a path, which is expected behaviour for this task.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install paper-collector - After installation, invoke the skill by name or use
/paper-collector - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.0.2
- 增加了对 `pinchtab` 和 `lightpanda` 第三方工具的优先调用说明，原有工具顺序更新。
- 工具优先级描述由具体工具名扩展为“第三方浏览器”等，更具适配性。
- 其他内容未作变更；整体运行逻辑与筛选流程保持一致。
v0.0.1
Initial release of paper-collector – a skill for focused academic paper collection based on user-specified research topics.
- Only runs when the research direction/theme is explicitly provided by the user; refuses requests that lack a clear topic.
- Retrieves and screens papers from journal websites, preprint platforms, and trusted academic sources; includes a paper only after reading its abstract for relevance.
- Supports optional “journal whitelist only” mode, adjustable paper count, and Markdown output with configurable file path.
- Enforces a one-month recency limit for all collected papers.
- Uses third-party tools (`agent-browser`, `tavily-search`) to find and verify papers, prioritizing official sources and abstracts.
- Returns structured output per paper: title, author, publication year, journal/source, link, relevance statement, and one-line abstract insight.
Metadata
Frequently Asked Questions
What is 文献采集?
按用户指定的研究方向检索并筛选学术论文：在期刊官网、预印本平台与学术来源中搜索，必须阅读摘要后才可纳入结果。适用于用户要求查找、收集、筛选或整理某一主题/研究问题/方向的文献；若未提供研究方向则不得执行。支持可选“仅限期刊清单”模式、可选论文数量覆盖、可选 Markdown 输出路径。 It is an AI Agent Skill for Claude Code / OpenClaw, with 224 downloads so far.
How do I install 文献采集?
Run "/install paper-collector" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is 文献采集 free?
Yes, 文献采集 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does 文献采集 support?
文献采集 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created 文献采集?
It is built and maintained by Tiandoufayale (@mxingchtongaelofficial2568); the current version is v0.0.2.
More Skills