文献助手
v1.0.0literature retrieval and single-pdf analysis for academic research. use when the user asks to find papers by topic, author, keyword, or latest papers in the...
⭐ 0· 40·0 current·0 all-time
by@lyt0302
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name and description (literature retrieval and single-PDF analysis) match the SKILL.md. However, the instructions assume the agent can perform web searches and fetch/confirm full-text PDFs and can parse uploaded PDFs; those capabilities are not declared in the skill metadata (no required tools or env). This is a functional gap rather than a security mismatch, but the user should confirm the agent has appropriate search/browsing and file-processing tools.
Instruction Scope
SKILL.md stays on task: it instructs how to build bilingual search strings, return structured search results, label availability, and to extract/analyze content from a single uploaded PDF. It does not direct the agent to read unrelated files, environment variables, or to transmit data to external endpoints beyond normal searching/fetching. It explicitly discourages fabrication.
Install Mechanism
No install spec and no code files — instruction-only skill (lowest install risk). Nothing is downloaded or written to disk by the skill itself.
Credentials
No required environment variables, credentials, or config paths are requested — proportional to the described functionality.
Persistence & Privilege
always is false and the skill does not request elevated or persistent privileges or modifications to other skills. Default autonomous invocation is allowed (normal for skills) but not combined with other concerning factors.
Assessment
This skill appears coherent and low-risk: it needs no secrets or installs. Before enabling, confirm the agent environment actually provides web-search/browsing and PDF parsing (so the skill can find papers and read uploaded PDFs). Be cautious about uploading PDFs that contain sensitive or unpublished data, and always manually verify any paper links, availability labels, and extracted facts (LLMs can still hallucinate despite the 'do not invent' instruction). If you need searches to come from a specific source (e.g., Google Scholar, Semantic Scholar, arXiv, institution subscriptions), verify or configure which tool or API the agent will use.Like a lobster shell, security has layers — review code before you run it.
latestvk976zvqc8qm61v0gjbhbh10k358429sf
License
MIT-0
Free to use, modify, and redistribute. No attribution required.