context7

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md and scripts explicitly fetch library documentation and code examples from public Context7 endpoints (https://api.context7.com and https://mcp.context7.com) and instruct the agent to read and use the returned docs to produce code, which clearly ingests untrusted third-party content that could contain indirect prompt-injection instructions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill performs runtime fetches to https://mcp.context7.com/mcp (and https://api.context7.com) to retrieve documentation and code examples which are parsed and injected into the agent's context/outputs, so external content can directly control prompts/instructions.