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). AWS credentials
come from your environment (AWS_REGION, AWS_ACCESS_KEY_ID…). BEDROCK_MODEL_ID is any model enabled in your
Bedrock account.
With the Python SDK
doconda.tools() is JSON Schema, the same format as a Strands tool spec:
With MCP
On Amazon Bedrock AgentCore
The agent is the same: deploy it to AgentCore Runtime like any Strands agent. Keep the Doconda key as a secret (DOCONDA_API_KEY), not in the code.
Tools
The agent gets the eight Doconda tools; see the tools for what each one does.doconda.tools() returns them with name, description, input_schema (JSON Schema) and run. 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: setDOCONDA_API_KEY(on AgentCore, from the secret), or passapi_keytoDoconda.- Bedrock rejects the model ID — The model must be enabled in your account and region, and support tool use.
- The download link stopped working — File links expire after 5 minutes; the agent calls
get_documentfor fresh ones.
Useful links
- Python SDK.
- Amazon Bedrock: the SDK tools with the Converse API.
- doconda-examples on GitHub: ready-to-run scripts.
- API reference.