Skillv0.1.0

ClawScan security

Google Scholar Search Skill · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignFeb 27, 2026, 1:18 AM

Verdict: benign
Confidence: medium
Model: gpt-5-mini
Summary: The skill's code, README, and SKILL.md are coherent with its stated purpose (Google Scholar scraping + scholarly author lookups); no hidden endpoints or credential requests were found, but some optional install instructions in the README (curl | sh and npx usage) are worth caution.
Guidance: This skill appears internally consistent with its stated goal of scraping Google Scholar and using the scholarly library for author data. Before installing: (1) avoid blindly running remote install commands (e.g., 'curl | sh' or arbitrary 'npx' commands) unless you trust the source — download and inspect scripts first; (2) install and run the Python code inside a virtualenv or container; (3) be aware that scraping Google Scholar can be blocked and may violate Google’s terms of service — consider using Semantic Scholar or PubMed APIs for stable access; (4) note references to Sci-Hub in the README (potential legal/copyright concerns) and decide if you want that linked functionality; (5) if you plan to use the 'scholarly' dependency, verify its behavior and network endpoints; and (6) verify the repository/author provenance (the README points at a GitHub user) before trusting remote install suggestions. These precautions would reduce risk if you choose to install and run this skill.

Review Dimensions

Purpose & Capability: okName and description match the included python implementation: requests + BeautifulSoup for scraping scholar.google.com and the scholarly library for author profiles. Required capabilities (network access and HTML parsing) align with the stated purpose; no unrelated credentials or binaries are requested.
Instruction Scope: okSKILL.md and README instruct only web scraping of Google Scholar and using the scholarly library for author metadata. The runtime instructions do not ask the agent to read unrelated local files, environment variables, or exfiltrate secrets. The SKILL.md explicitly notes Google Scholar has no official API and that scraping may be blocked.
Install Mechanism: noteRegistry has no formal install spec (instruction-only), but the README provides multiple optional install methods. Two items to watch: (1) a recommended 'curl -LsSf https://astral.sh/uv/install.sh | sh' to install 'uv' — piping a remote shell script to sh carries risk if you don't trust that host; (2) the one-click 'npx skills add https://github.com/...' suggestion may execute tooling that fetches remote code. The Python dependencies themselves (requests, beautifulsoup4, scholarly) are standard and declared in requirements.txt.
Credentials: okThe skill declares no required environment variables, no secrets, and no config paths. The code does not attempt to access system credentials or unrelated environment values. Network access to scholar.google.com is required and appropriate for the stated purpose.
Persistence & Privilege: okThe skill does not request always:true or any elevated/always-present privilege. It does not modify other skills or system-wide settings. It appears to operate as a user-invoked CLI/skill.