About Doconda
Doconda's MCP server lets AI agents create, review, and edit Word, PowerPoint, Excel, and PDF documents. It provides editable files along with a PDF and page previews for review before delivery.
Key features
- 01Create Word, PowerPoint, Excel, and PDF files
- 02Review and edit documents
- 03Upload and retrieve files
- 04Read artifact data and reports
- 05Provide PDF and page previews
Use cases
- 01Generate an Excel event budget
- 02Review and revise a Word document
- 03Create a presentation or PDF from an agent request
MCP support
Exposes document creation, upload, review, editing, retrieval, reporting, and artifact-reading tools to MCP clients. Available as a remote Streamable HTTP server or a local stdio server.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://api.eu.doconda.com/mcpclaude mcp add --transport http com-doconda https://api.eu.doconda.com/mcp{"mcpServers":{"com-doconda":{"type":"http","url":"https://api.eu.doconda.com/mcp"}}}Open in Cursor{"servers":{"com-doconda":{"type":"http","url":"https://api.eu.doconda.com/mcp"}}}Open in VS Code{"mcpServers":{"com-doconda":{"type":"http","url":"https://api.eu.doconda.com/mcp"}}}{"mcpServers":{"com-doconda":{"serverUrl":"https://api.eu.doconda.com/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
https://api.eu.doconda.com/mcpAuth requiredTools
8| Parameter | Type | Required | Description |
|---|---|---|---|
| data | enum | No | Artifacts only. shared (default): everyone with the link sees the same data (polls, team lists). local: each visitor keeps their own in their browser (calculators, personal checklists). With retention none, only local.shared | local |
| name | string | No | Optional: a name to find it later, e.g. 'Contrato piso Malasaña'. It also names the files. |
| style | string | No | Optional: the look in plain words, e.g. 'blue centred headings, Lora font, page numbers, table of contents'. |
| format | enum | No | docx → Word (+ PDF); pdf → only the PDF; pptx → PowerPoint (+ PDF); xlsx → Excel (+ PDF); artifact → web page. If the user didn't say, leave it out and Doconda decides from the prompt.docx | pdf | pptx | xlsx | artifact |
| prompt | string | No | What to write, in the user's words (topic, audience, length, tone). Formatting wishes too, e.g. 'blue centred headings, with a cover page'. |
| markdown | string | No | Content already written, in Markdown (instead of prompt). For an artifact, the content its page shows. |
| retention | enum | No | Optional: how long Doconda keeps it once finished (none = 15 minutes). Without it, the project's retention (30 days unless changed). The user should save the file before then.none | 1d | 7d | 30d | 90d |
| source_file_ids | array | No | Uploaded files (upload_file) or documents already made (their id, from list_documents): documents to base the text on (up to 5, with prompt) and pictures to put in the document (up to 10; with markdown, refer to them as , the first picture). |
Details
From the project README@doconda/mcp
MCP server for Doconda: give your AI agent tools to create, review and edit Word, PowerPoint, Excel and PDF documents. Editable files, reviewed before delivery, with a PDF and a preview of each page.
Remote server (for your app's agent)
No install: https://api.eu.doconda.com/mcp (Streamable HTTP, header Authorization: Bearer ak_eu_…). Any agent
framework or model API that speaks MCP can use it. With the OpenAI Responses API:
from openai import OpenAI
response = OpenAI().responses.create(
model="gpt-5",
input="Make an Excel with the event budget",
tools=[{
"type": "mcp",
"server_label": "doconda",
"server_url": "https://api.eu.doconda.com/mcp",
"headers": {"Authorization": "Bearer ak_eu_…"},
"require_approval": "never",
}],
)
print(response.output_text)
Messages (errors, report texts) come in English; the remote server follows Accept-Language, so send
Accept-Language: es for Spanish.
Not using MCP? The SDKs return the same tools for any model with tool calling: @doconda/sdk · doconda.
Local server (stdio)
With Claude Code:
claude mcp add --transport stdio doconda --env DOCONDA_API_KEY=ak_eu_… -- npx -y @doconda/mcp
For other MCP clients, such as Claude Desktop:
{
"mcpServers": {
"doconda": {
"command": "npx",
"args": ["-y", "@doconda/mcp"],
"env": { "DOCONDA_API_KEY": "ak_eu_…" }
}
}
}
Tools
create_document, upload_file, review_document, edit_document, list_documents, get_document, get_report,
read_artifact_data.
Docs: https://docs.doconda.com/en/agents/mcp · Examples: https://github.com/B-C-Software/doconda-examples · Get an API key: https://doconda.com
Development
Node 22.18 or later:
npm install
npm run build
npm test
The tests need no API key and make no network calls. This repository is published from Doconda's main repository: open issues and pull requests here, and accepted changes are applied upstream and arrive with the next release.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.