Literature Reviewer Skill

Security checks across malware telemetry and agentic risk

Overview

The skill is purpose-aligned for literature reviews, but users should notice its browser automation, possible logged-in database use, local session files, and unpinned GitHub install guidance.

This skill appears suitable for literature-review tasks. Before installing or using it, prefer the reviewed package over an unpinned GitHub clone, supervise browser automation on academic sites, avoid unapproved bulk downloads, and remember that session folders may retain your research topic and collected paper metadata.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

#
ASI02: Tool Misuse and Exploitation
Low
What this means

The agent may open academic databases, fill search forms, and scrape result metadata automatically.

Why it was flagged

The skill directs the agent to automate browser actions on academic database sites. This is central to the stated purpose, but it can generate web traffic and interact with third-party sites on the user's behalf.

Skill content
使用 `browser_navigate` 和 `browser_fill_form` 等工具访问数据库网页,提取文献信息。
Recommendation

Review requested searches and stop the workflow if a site asks for captcha, payment, login, or actions beyond metadata retrieval.

#
ASI03: Identity and Privilege Abuse
Low
What this means

If used for downloads, the skill could consume institutional or paid database access through the browser session.

Why it was flagged

The reference guide discusses bulk PDF downloading that may depend on a logged-in account or paid access. This is related to literature review, but users should explicitly approve any account-backed downloads.

Skill content
CNKI 对批量下载有限制
- 需要登录账号
- 部分文献需要购买权限
Recommendation

Only allow downloads you intend, confirm licensing or payment implications, and prefer metadata-only searches unless full text is explicitly needed.

#
ASI04: Agentic Supply Chain Vulnerabilities
Low
What this means

Installing from the live repository could fetch code or instructions different from the reviewed package.

Why it was flagged

The README recommends installing from an external GitHub repository without pinning a commit or release. This is common but leaves the installed content dependent on the repository state at install time.

Skill content
git clone https://github.com/stephenlzc/AI-Powered-Literature-Review-Skills.git literature-reviewer-skill
Recommendation

Install from the reviewed registry package or pin and inspect a specific Git commit before cloning into a skills directory.

#
ASI06: Memory and Context Poisoning
Info
What this means

Your research topic and collected literature metadata may be saved locally between sessions.

Why it was flagged

The skill persists search results, abstracts, analysis, and final outputs in session files. This is expected for resumable literature review, but it means retrieved web content and user topics remain available for later reuse.

Skill content
sessions/{YYYYMMDD}_{topic_short}/ ... papers_raw.json ... papers_analysis.json ... literature_review.md
Recommendation

Use a project-specific workspace, review generated files before reuse, and delete session folders if the topic or notes are sensitive.