process-meeting-notes

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

• Third-party content exposure detected (high risk: 0.95). Outsider free text can enter the LLM context via Fireflies MCP transcript retrieval (mcp__fireflies__fireflies_get_transcript / mcp__fireflies__fireflies_get_summary) in the meeting-processing workflows, since Fireflies transcripts are authored by meeting participants other than the operating user.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires an MCP server invoked via npx (see the MCP config: "command": "npx", "args": ["-y", "@anthropic/fireflies-mcp"]), which will fetch and execute remote npm package code at runtime to provide the mcp__fireflies__* commands used by the workflows.