claude-ai

Anthropic开发的安全AI助手

Security Scan

The skill is mostly harmless and coherent with its stated function (an encyclopedia entry about Claude AI), but it claims to be 'Anthropic-developed' while having no source/homepage and a different listed author — this inconsistency and the skill's vague instruction to perform unrestricted web searches merit caution.

Purpose & Capability

The skill's name/description say it's an Anthropic-developed security AI assistant, but the SKILL.md content is actually an encyclopedia-style info provider about Claude AI. The registry metadata lists a different author/owner and there is no source or homepage — possible misattribution or misleading branding, though the capabilities requested (none) are proportionate to an informational skill.

Instruction Scope

Instructions are instruction-only and limited to 'Uses web search to fetch real-time, comprehensive information from authoritative sources.' No file, env, or credential access is requested. However, 'web search' is vague and grants the agent broad discretion over external sites to contact and what to fetch.

Install Mechanism

No install spec and no code files — lowest-risk installation model (instruction-only).

Credentials

No environment variables, binaries, or config paths are required — proportional for an informational skill.

Persistence & Privilege

always is false and model invocation is allowed (the platform default). This is normal; there are no elevated privileges or requests for persistent credentials.

What to consider before installing

This skill itself requests no secrets or installs and appears to only fetch web information, but exercise caution before installing: the description claims 'Anthropic-developed' while the owner/author and lack of homepage/source suggest it is not official — that could be misleading branding. If you install it, review and control the agent's web-browsing permissions (or disable autonomous invocation) so the skill can't fetch arbitrary external pages without your consent. Prefer official or well-documented skills for anything that claims to be from a particular vendor, and consider asking the publisher for provenance (source code or homepage) before trusting it.

Anthropic开发的安全AI助手

What This Skill Does

This skill provides comprehensive information about Claude AI:

1. Overview — What it is, history, key facts
2. Latest Updates — Recent news, developments, and announcements
3. Key Data — Statistics, rankings, records, and metrics
4. Resources — Official websites, useful links, and references
5. Related Topics — Connected entities and recommendations

Example Queries

• "Tell me about Claude"
• "Latest news on Claude"
• "Claude key facts"

How It Works

Uses web search to fetch real-time, comprehensive information from authoritative sources.

Author

Created by hanxueyuan | Part of the Encyclopedia project | License: MIT