Open Data Framework CLI (odf)

Scaffolds and runs Open Data Framework projects, and serves them through a UI, an MCP server and a chat window. Depends on opendataframework for the core abstractions (Entity, Repository, Component, Service, Task, Pipeline, Agent, Function, Layer, Context, Project) — see that package for those.

Documentation: https://opendataframework.github.io/odf/

Install

pip install odf                # CLI + scaffolding only
pip install "odf[ui]"          # + UI dev server
pip install "odf[mcp]"         # + MCP server
pip install "odf[ui,chat]"     # + chat window (served by the UI)

odf run starts the UI by default, so most projects want at least odf[ui].

Quickstart

pip install "odf[ui]"
odf init my-project            # or --template data-analytics / data-engineering / data-science / ml / research
cd my-project
odf run                        # UI at http://127.0.0.1:4747

With pip install "odf[ui,mcp,chat]", odf run --mcp --chat adds the MCP server at http://127.0.0.1:4748/mcp and a chat window. The same from Python:

from odf.server import Server

server = Server.from_config("config.toml")
server.start(ui=True, mcp=True, chat=True)
server.wait()  # blocks until Ctrl+C/SIGTERM

What's included

• CLI — odf init scaffolds a project from a template; odf run starts it.
• UI — the topology of every component with its actions, a view of each repository's data (table, map, timeseries, image, video, audio, document), and labelling of that data into a plain annotations.json.
• MCP server — the same actions as MCP tools, plus each agent, function and component's own tools.
• Chat — a chat window in the UI, backed by a local Ollama model, that acts on the project through the MCP tools.

See the documentation for each, and examples/ for 22 small projects, one concept each.