ClawHub
Back to skill

Security audit

Context7 MCP Skill

Security checks for vulnerabilities and agentic risk

Overview

The skill is a straightforward Context7 documentation lookup helper with disclosed network use and a local command link setup step.

Before installing, be aware that documentation queries go to Context7 and that the first run may create a local context7-mcp-cli link through uxc. Avoid sending private code or secrets in documentation queries unless you are comfortable with that external service receiving them.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Tp4

High
Category
MCP Tool Poisoning
Confidence
96% confidence
Finding
The declared purpose describes an operational skill for retrieving current library documentation through Context7 MCP. However, the actual code does not perform any documentation lookup, MCP interaction, network access, or code example retrieval. Instead, it validates that the skill's local metadata and docs contain required strings and avoid forbidden patterns. This is a materially different primary purpose: build/CI-style validation of a skill package rather than end-user documentation querying. The references to Context7 tools and endpoint are only checked as text content in files, not used to provide the declared functionality.

Static analysis

No suspicious patterns detected.