blueprint-mcp-diagram-generation

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly shows the agent fetching external content (e.g., hubspot_get_deal in "Pattern 2: Multi-Tool Integration (HubSpot + Diagram)", gdrive_read_file in "Integration Examples", and github_list_files) and embedding that untrusted, user-generated third-party content into prompts and follow-up actions, so those sources can materially influence tool use and enable indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill requires runtime calls to the Arcade MCP gateway (e.g. https://api.arcade.dev/gateway/YOUR_GATEWAY_ID) to submit prompts via start_diagram_job which causes remote diagram-generation code to execute and thus externally controls agent behavior.