google-genai in Python, or @ai-sdk/google-vertex and
@google/genai in TypeScript.
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). Google Cloud
credentials come from your environment (gcloud auth application-default login or a service account), like the rest
of your code.
With Python
res.text includes the spreadsheet’s download links.
With the AI SDK
With @google/genai
vertexai: true:
Tools
The agent gets the eight Doconda tools; see the tools for what each one does.doconda.tools() returns them with name, description, a JSON Schema (inputSchema in TypeScript, input_schema
in Python) and run; Gemini takes the schema as is in parametersJsonSchema (parameters_json_schema in Python). A
failed call returns { "error": "<code>", "message": "…" } instead of raising, so the model can correct itself.
Troubleshooting
Doconda: pass api_key or set DOCONDA_API_KEY.— The SDK found no key (in TypeScript:Doconda: pass { apiKey } or set DOCONDA_API_KEY.). SetDOCONDA_API_KEY, or pass the key toDoconda.- Vertex AI can’t find the model — Check
projectandlocation: the model must be available in that region. - 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
- Python SDK and TypeScript SDK.
- Gemini and Google ADK.
- doconda-examples on GitHub: ready-to-run scripts.
- API reference.