Research Paper Monitor
v1.0.2科研文献智能监测与摘要推送系统。自动监测多个学术信源（arXiv、PubMed、CNKI等），根据用户关注的领域和关键词采集最新论文，生成中文摘要并推送。适用于需要跟踪学术前沿的科研工作者、研究生、教师等。使用场景包括：(1) 定时监测特定研究领域的最新论文，(2) 根据关键词筛选高相关度论文，(3) 自动生成论...
⭐ 3· 296·0 current·0 all-time
by孙金刚@wiltonmotta
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description match the included scripts (config.py, monitor.py, search_papers.py, generate_digest.py). The code implements multi-source paper search (arXiv stubbed, PubMed/GoogleScholar stubs), scoring, local archiving, and optional push notifications — all consistent with the stated purpose. No unrelated cloud credentials or binaries are requested.
Instruction Scope
Runtime instructions are scoped to reading/writing local config/index files under ~/.openclaw/research-monitor, calling public academic APIs (arXiv) or web searches, generating summaries, and optionally sending notifications. The core monitor.py uses only urllib/ET for arXiv calls and writes local markdown/JSON files. The documentation includes optional examples (email via SMTP, WeCom, Zotero integration) that, if copied in, would require credentials and enable external posting — these are optional extensions, not mandatory behavior in the provided core scripts.
Install Mechanism
No install spec is provided; this is effectively a collection of local Python scripts. Nothing is downloaded or extracted during install by the skill itself. That lowers risk compared to arbitrary remote downloads.
Credentials
The skill declares no required environment variables or primary credentials. It may prompt for an optional Feishu webhook URL (configurable) and the docs show optional integrations (SMTP, Zotero API key) which would require credentials only if the user enables them. There are no hidden or unexplained secret requests in the core code.
Persistence & Privilege
The skill writes data only to its own directories under ~/.openclaw/research-monitor and ~/.openclaw/research-papers and updates its own index file. always is false and it does not modify other skills or system-wide settings.
Assessment
This package appears to do what it says: local collection, scoring, Chinese summarization templates, and local archiving. Before running: (1) review monitor.py and config.py to confirm the target directories and that you’re comfortable with files being written under ~/.openclaw; (2) only provide webhooks, SMTP, or Zotero API keys if you intend to enable those optional integrations — those will send data off your machine; (3) run the scripts under a regular user (not root) and test with a minimal configuration (few keywords, disabled notifications) to observe network requests and outputs; (4) if you need stronger guarantees, inspect or run the code in a sandbox/container and consider replacing the template summarization with your preferred local/remote LLM integration after reviewing its privacy implications.Like a lobster shell, security has layers — review code before you run it.
latestvk975kqkk4kxhd1hd9yckf0sry9839jj3
License
MIT-0
Free to use, modify, and redistribute. No attribution required.