Problemy Infakt MCP z Google Gemini

Próbuję połączyć się z Infakt MCP poprzez Agenta AI w n8n.
Przy użyciu modeli OpenAI lub Anthropic wszystko działa w porządku.
Niestety wygląda na to, że Google Gemini ma inne wymagania co do formatu listy tooli zwracanych przez MCP. 

[GoogleGenerativeAI Error]: Error fetching from https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-flash:generateContent: [400 Bad Request] Invalid JSON payload received. Unknown name "type" at 'tools[0].function_declarations[1].parameters.properties[0].value.properties[3].value': Proto field is not repeating, cannot start list. Invalid JSON payload received. Unknown name "type" at 'tools[0].function_declarations[1].parameters.properties[0].value.properties[4].value': Proto field is not repeating, cannot start list. Invalid JSON payload received. Unknown name "type" at 'tools[0].function_declarations[1].parameters.properties[0].value.properties[5].value': Proto field is not repeating, cannot start list. Invalid JSON payload received. Unknown name "type" at 'tools[0].function_declarations[2].parameters.properties[0].value.properties[2].value': Proto field is not repeating, cannot start list. 

Problemy na temat formatowania listy tooli z innymi platformami opisane poniżej: 
https://discuss.ai.google.dev/t/gemini-rejects-tool-json-schema-with-array-of-types/4580
https://community.baserow.io/t/n8n-connecting-ai-agent-to-google-gemini-and-mcp-server-leads-to-error/10071
https://community.fibery.io/t/some-new-mcp-tools-are-breaking-ai-models/10606