Biztory Tableau Knowledge MCP

biztory-tableau-knowledge-mcp

Put your Tableau knowledge graph to work in any AI assistant. Search it by meaning, trace lineage and impact, and enrich it with business context.

biztory-tableau-knowledge-mcp turns your site's Tableau Knowledge Graph API into twenty-three MCP tools, so an AI assistant (Claude, Cursor, VS Code, Cline, and others) can work with the graph in plain language:

• _"Which fields mean revenue, and where do they come from?"_
• _"If the Superstore data source changes, which workbooks and dashboards are affected?"_
• _"What business context is attached to Profit? Add: 'Profit is after returns, in EUR.'"_
• _"How healthy is our graph, and what are the five highest-impact fixes?"_
• _"Export Superstore to Apache Ossie so we can port it to Snowflake."_
• _"Write the Business Preferences for the P&L dashboard."_
• _"Add every Superstore data source on the site to the Sales graph."_
• _"Ask Tableau's agent how many data sources the graph has, and check its answer."_

It ships nine read tools (semantic search, node lookup, edges, lineage, impact, sources, semantic contexts, graph health, and a connection probe) and four write tools for semantic contexts, the business statements that make the graph useful to people and to AI (one turns a document into context). Five tools manage the graph itself: add and remove data sources and workbooks, create and update graphs, and ask Tableau's own agent a question to test its answers. Those go through the API behind the Tableau Knowledge web app, which Tableau doesn't document. Four more tools port semantic layers through Apache Ossie (formerly Open Semantic Interchange, OSI), the open standard that Snowflake, dbt, Power BI, Databricks, Tableau Next and others convert to and from. And one checks Business Preferences files line by line against the fields the agent will actually see.

See it in action

Replays of real tool results from a test site, with site-specific names removed. Regenerate with docs/images/render_demo_gifs.py .

Why it's different

• Ask by meaning, not by id. Every read tool takes a node id *or* a description ("the net sales field"). Descriptions resolve through the graph's own semantic search, and the answer says which node it picked, with the match score and runners-up, so a wrong guess is visible instead of silent.
• Answers, not payloads. impact_analysis groups and counts everything downstream by type; lineage adds node and edge counts; graph_health caps hundreds of suggestions to the top ones while keeping every total.
• Errors tell the model what to do next. Tableau's error codes become concrete steps (site_not_entitled → an admin must enable Knowledge; no_active_graph → pass a graph_id; no embedder → use exact ids), with the trace_id kept for support. Rate limits are retried with Retry-After.
• Writes you can trust. Only semantic contexts and a graph's sources can be written (the latter through Tableau Knowledge's undocumented web-app API, which can change without notice). Write tools take exact ids (a fuzzy match never decides where a write lands), validate before sending, carry honest MCP annotations so clients ask first, and delete_semantic_context returns what it deleted.
• Port your semantic layer. export_to_ossie turns a published data source into a validated Ossie model, with Tableau calcs translated to portable SQL wherever that is exact and a loss report for everything else. import_from_ossie brings descriptions, synonyms and definitions from any Ossie model back into Tableau Knowledge, with a reviewable plan first.
• Tiny and hardened. Stdio only (no inbound port), secrets from env only, HTTPS enforced, redirects refused, ids encoded so they can never reach another endpoint, fuzzed parsers, and two runtime dependencies: the MCP SDK and PyYAML (stdlib urllib for HTTP).

Requirements

Tableau Knowledge is a Tableau Cloud feature (September 2026 and later, Tableau+ license); it is not available on Tableau Server. Any user can read the graph, filtered by their content permissions. Writing semantic contexts needs the Creator role.

Quickstart

You need uv (curl -LsSf https://astral.sh/uv/install.sh | sh) and a Tableau Personal Access Token.

Claude Code, one line:

claude mcp add tableau-knowledge \

-e TABLEAU_SERVER=https://10ax.online.tableau.com \ -e TABLEAU_SITE_CONTENT_URL=YourSite \ -e TABLEAU_PAT_NAME=my-token \ -e TABLEAU_PAT_SECRET=the-full-secret \ -- uvx biztory-tableau-knowledge-mcp

That's all. uvx fetches the package from PyPI and runs it in an isolated environment; nothing to clone or install. Then ask: _"Use server_info: what graphs does my site have?"_

Configuration

All configuration is via environment variables (set them in your client's env block, never on the command line).

¹ Provide a PAT (TABLEAU_PAT_NAME + TABLEAU_PAT_SECRET) or TABLEAU_AUTH_TOKEN.

Claude Desktop

Edit claude_desktop_config.json (macOS: ~/Library/Application Support/Claude/, Windows: %APPDATA%\Claude\):

{

"mcpServers": { "tableau-knowledge": { "command": "uvx", "args": ["biztory-tableau-knowledge-mcp"], "env": { "TABLEAU_SERVER": "https://10ax.online.tableau.com", "TABLEAU_SITE_CONTENT_URL": "YourSite", "TABLEAU_PAT_NAME": "my-token", "TABLEAU_PAT_SECRET": "the-full-secret" } } } }

Fully quit and reopen Claude Desktop, then check the tools menu.

Cursor · VS Code · Cline · Windsurf

Use the Add to Cursor / Add to VS Code buttons above and fill in the env values, or add the same block as for Claude Desktop to:

• Cursor: ~/.cursor/mcp.json (global) or .cursor/mcp.json (project).
• VS Code: .vscode/mcp.json (the top-level key there is servers instead of mcpServers).
• Cline: the MCP Servers panel, then *Configure*, into cline_mcp_settings.json.
• Windsurf: ~/.codeium/windsurf/mcp_config.json.

On Windows, if uvx isn't found by the GUI app, use its absolute path (e.g. %USERPROFILE%\.local\bin\uvx.exe).

Tools

node arguments accept a node id or a plain-language description. Every graph-scoped tool also takes an optional `grap