Apple Docs

• Outcome: Review
• Latest audit: May 27, 2026, 8:57 PM
• Version:
  1.0.0

Overview

This looks like a legitimate Apple documentation lookup tool, but one command can fetch arbitrary web URLs beyond Apple documentation.

Install only if you are comfortable with a read-only docs tool that makes outbound requests. Prefer using it for Apple Developer documentation paths and WWDC lookups; be cautious about letting an agent pass full URLs to the doc command because it can retrieve non-Apple or internal network content if the runtime can reach it.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (2)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 88% confidence
• Finding: The skill explicitly states it uses native fetch and references an external MCP server plus Apple documentation endpoints, which indicates network access is part of its behavior. Having effective network capability without declared permissions weakens transparency and policy enforcement, making it harder for a host system or user to assess data exfiltration risk or restrict outbound requests.

Context-Inappropriate Capability

Medium

• Confidence: 95% confidence
• Finding: The `doc` command accepts any user-supplied `http`/`https` URL and then fetches it directly, rather than restricting requests to Apple Developer documentation domains and paths. This creates an SSRF-style primitive in the skill context: an agent or user can be induced to retrieve arbitrary remote content, including internal services if network access is available, and then print returned data to output.

VirusTotal

66/66 vendors flagged this skill as clean.