Paper Compare
PassAudited by ClawScan on May 1, 2026.
Overview
This is an instruction-only academic paper comparison skill whose disclosed web, PDF, and history use appears aligned with its purpose.
This skill appears safe to install as an instruction-only research helper. Before using it, be aware that it may search the web or academic APIs for DOIs, URLs, and queries, may extract text from PDFs, and may refer to prior comparison history if available.
Findings (2)
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.
Paper DOIs, URLs, or search terms may be looked up through external academic/search services as part of the comparison.
The skill instructs the agent to use web/API retrieval tools to find paper metadata and content. This is expected for comparing papers, and search confirmation is explicitly requested.
DOI ──→ Fetch via crossref/semantic scholar ... URL ──→ Fetch via web_fetch ... Search query ──→ Search first ... Use web_search, present top 3, CONFIRM before proceeding
Use this with papers or queries you are comfortable sending to web and academic metadata services; review search results before proceeding.
Previous paper comparisons or saved user goals could be reused in later outputs, which may be undesirable for private research topics or unrelated projects.
The skill may retrieve prior comparison history, which can help with continuity but also means old context could influence new analyses.
History | memory_search | Prior comparisons
If privacy or separation between projects matters, ask the agent not to use prior history and verify that conclusions are based only on the current papers.