Paper_Briefing_Assistant
PassAudited by ClawScan on May 1, 2026.
Overview
This is a coherent instruction-only literature survey workflow, with disclosed web research and user checkpoints; the main thing to notice is optional memory use for prior research runs.
This skill appears safe for normal academic literature surveys. Before installing, be aware that it will perform external web searches and may store the search strategy in memory for reproducibility; avoid confidential research topics or ask the agent not to use memory if that matters.
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.
Your research topic and search terms may be used in external web queries during the survey.
The skill clearly instructs the agent to use web retrieval tools and possibly parallel sessions. This is central to the literature-survey purpose, but it means user topics and generated keywords may be sent to external academic/search services.
Use available tools to query, in sequence or in parallel as appropriate: Google Scholar, Semantic Scholar, arXiv, IEEE Xplore, and other relevant academic platforms.
Use non-sensitive topic descriptions when possible, and rely on the built-in checkpoints to confirm the retrieval scope before deeper analysis.
Research topics, keywords, and prior-run strategy may be retained and reused in future sessions.
The skill uses memory to store and reuse research strategies across runs. This is coherent with reproducibility, but persistent memory can retain sensitive research interests or allow stale/incorrect prior context to influence later work.
Record the full strategy in memory (keywords, time range, database priority) for the Method Appendix ... memory_search / memory_get | Cross-reference prior runs
If the topic is confidential, ask the agent not to use persistent memory for the task or clear relevant memory after the survey; verify any reused prior results.