An MCP server exposing data science tools (plotting, stats, data exploration) to any LLM client.
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"ds-mcp-server","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# ds-mcp-server `ds-mcp-server` packages a FastMCP server with data science, plotting, statistics, system, and web tools, plus interactive CLI clients for OpenAI-compatible providers and Anthropic Claude. ## What's in the box After you `pip install ds-mcp-server`, three commands are available: | Command | What it is | When to use it | |---|---|---| | **`ds-mcp-webui`** | Browser chat UI | You want to chat and see plots in your browser. **Start here.** | | **`ds-mcp-client`** | Interactive terminal chat | You prefer the CLI. Same features as the web UI, minus inline plot rendering. | | **`ds-mcp-server`** | The MCP server itself | You are configuring an **external** MCP client (Claude Desktop, LM Studio, Cursor, etc.) to launch it. **Do not run this by hand** — it will look "frozen" because it's silently waiting for MCP protocol messages on stdin. | > **In short:** for humans → `ds-mcp-webui` or `ds-mcp-client`. > For MCP clients configured with a `command` field → `ds-mcp-server`. > ⚠️ **Intended use: local, single-user.** > The web UI (`ds-mcp-webui`) and terminal client (`ds-mcp-client`) are designed > to be run **locally by one person on their own machine** — like Jupyter, > Streamlit, or a local Ollama UI. They are **not** multi-tenant and have no > authentication, per-user isolation, or rate limiting: a single shared MCP > session, global settings, and one set of API keys are shared by every > connection. Some tools (the opt-in **system tools** and **unrestricted > custom-plot exec**) can execute code with your privileges. > > **Do not expose these servers to the public internet or untrusted users or > networks.** Bind them to `localhost` (the default), and if you need remote > access, put them behind your own authenticated tunnel/VPN. Enabling the > dangerous toggles on a machine reachable by others is a remote-code-execution > risk. See [Optional system tools](#optional-system-tools) for details. ## Installation Install from PyPI: ```bash pip instal…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.