portfolio-rebalancer

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

• Third-party content exposure detected (high risk: 0.75). The required runtime path calls the public planfi MCP tool (analyze_rebalancing_trades and optionally generate_financial_plan) which returns structured outputs including trades[], drift[], and assumed_defaults[]; because the MCP is an external service, its returned text/fields are outsider-authored content that the agent will place into the LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). This skill explicitly requires calling the planfi MCP at runtime (https://ai.planfi.app/mcp/free and https://ai.planfi.app/mcp) to run the analyze_rebalancing_trades/generate_financial_plan tools, so remote server-side code executes and directly controls the agent's outputs.