It Searching

Provides up-to-date tech news and AI research by scraping major AI company blogs and retrieving relevant papers from arXiv.

Security Scan

An instruction-only skill that scrapes public AI blogs and arXiv pages; its requests and instructions are consistent with its stated purpose and it does not ask for credentials or install anything.

Purpose & Capability

The skill's name/description (fetch tech news and arXiv papers) matches the instructions which use url_scraping and arxiv_search tools to read public blogs and arXiv pages. Minor operational mismatch: the SKILL.md assumes availability of 'url_scraping' and 'arxiv_search' tools but the registry metadata does not declare required tools or integrations — this is an implementation/packaging omission, not a security discrepancy.

Instruction Scope

Instructions are narrowly scoped to scraping listed company blogs and arXiv pages and to using arxiv_search only for targeted queries. The content does not instruct reading local files, environment variables, or unrelated system state. It does not mention respecting robots.txt, rate limits, or site policies (operational guidance missing).

Install Mechanism

No install spec and no code files — instruction-only skill, so nothing is written to disk and there are no third-party packages pulled in.

Credentials

The skill requests no environment variables, credentials, or config paths. That is proportionate for a web-scraping/search skill.

Persistence & Privilege

always:false and standard autonomous invocation settings. The skill does not request persistent system-wide privileges or modify other skills' configs.

Assessment

This skill is instruction-only and appears coherent with its purpose. Before installing, confirm that your agent environment provides the 'url_scraping' and 'arxiv_search' tools the skill expects. Be aware the skill will fetch content from external websites (public blogs and arXiv); ensure that scraping those sites is acceptable for your use-case and that the scraping tool respects robots.txt and rate limits. Because no credentials are requested, there is low direct credential-exfiltration risk, but scraped pages can contain misleading or malicious content—validate important facts against original sources. If you need more control, request the publisher declare required tools or add explicit limits (e.g., domains, rate limits, robots.txt compliance) in the SKILL.md.

Overview

This skill provides specialized capabilities for it searching.

Instructions

Current Date: $DATE$. You are a Tech Search Agent with access to url_scraping and arxiv_search tools. Known tech company blog URLs:- OpenAI: https://openai.com/blog/- Google AI: https://ai.googleblog.com/- DeepMind: https://deepmind.com/blog/- Meta AI: https://ai.facebook.com/blog/- Microsoft Research: https://www.microsoft.com/en-us/research/blog/- Anthropic: https://www.anthropic.com/news- Hugging Face: https://huggingface.co/blog- NVIDIA: https://blogs.nvidia.com/- ArXiv: https://arxiv.org/Tool Usage Rules:1. For queries like "today’s AI papers" or "recent papers", use url_scraping to read ArXiv pages (start with https://arxiv.org/list/cs.AI/recent or category pages), NOT arxiv_search2. Use arxiv_search only for specific paper titles, authors, or technical concepts3. For company news, directly use url_scraping on their known blog URLsAlways provide accurate, up-to-date information by reading the actual web content.

Usage Notes

• This skill is based on the IT_Searching agent configuration
• Template variables (if any) like $DATE$, $SESSION_GROUP_ID$ may require runtime substitution
• Follow the instructions and guidelines provided in the content above