@google/genai; its experimental MCP support works with the remote
MCP server.
Capabilities
- Create a Word, PowerPoint, Excel or PDF file, or an artifact, from a request or from Markdown the agent wrote.
- Edit a file: only what is asked changes, and
get_reportlists every change. - Convert a file to another format with no AI, for example any file to Markdown so the agent can read it.
- Read what people saved in an artifact, such as poll votes.
Setup
This integration uses a Doconda API key inDOCONDA_API_KEY (from API keys). GoogleGenAI
reads your Gemini key from GEMINI_API_KEY. For Python, use the google-genai loop on the Vertex AI
page.
With the SDK tools
res.text includes the spreadsheet’s download links.
With MCP
With MCP (experimental in @google/genai)
With MCP (experimental in @google/genai)
Tools
The agent gets the eight Doconda tools; see the tools for what each one does.doconda.tools() returns them with name, description, inputSchema (JSON Schema, which Gemini takes as is in
parametersJsonSchema) and run. A failed call returns { "error": "<code>", "message": "…" } instead of throwing,
so the model can correct itself.
Troubleshooting
Doconda: pass { apiKey } or set DOCONDA_API_KEY.— The SDK found no key: setDOCONDA_API_KEY, or pass{ apiKey }toDoconda.- A tool result has
"error": "insufficient_balance"— The project’s balance can’t cover the document: top up in the dashboard, or ask forquality: "fast". - The download link stopped working — File links expire after 5 minutes; the agent calls
get_documentfor fresh ones.
Useful links
- TypeScript SDK.
- Google Vertex AI: the same loop on Vertex AI, and in Python.
- doconda-examples on GitHub: ready-to-run scripts.
- API reference.