sales-read-ai

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to connect to Read.ai's MCP endpoint (https://api.read.ai/mcp/) and to consume webhook/REST payloads containing full meeting transcripts and synced content from third‑party sources (Gmail, Slack, Google Drive, etc.) as part of its workflow (see SKILL.md and references/platform-guide.md), meaning untrusted user-generated content is read and used to drive actions and decisions and could embed indirect prompt-injection instructions.