CSDK Docs MCP Server
Enables TF-IDF search and structured browsing of Sisense Compose SDK documentation for React, Vue, and Angular, with support for framework filtering and code context to find relevant code examples.
README
CSDK Docs MCP Server
An MCP server that provides both structured browsing and TF-IDF search over the Sisense Compose SDK documentation. Works with Claude Code, Claude Desktop, or any MCP client.
What it does
Tools
| Tool | Description |
|---|---|
search_csdk_docs |
TF-IDF search over 1,700+ doc chunks. Returns the most relevant documentation for any query. |
browse_csdk_docs |
Browse docs by category (guides, react, vue, angular, data, design). List files in a category or read a specific file. |
list_csdk_topics |
Lists all available categories, files, and their descriptions. |
Resources
The server also exposes structured resources that MCP clients can browse directly:
| Resource | URI | Description |
|---|---|---|
| Full index | csdk://index |
Complete documentation index across all categories |
| Category index | csdk://{category} |
List files in a category (e.g., csdk://react, csdk://guides) |
| Doc file | csdk://{category}/{file} |
Read a specific doc file (e.g., csdk://react/charts.md) |
Two ways to access docs
- Search (
search_csdk_docs) — when you have a question and need the most relevant chunks. Uses TF-IDF scoring across 1,700+ pre-indexed chunks. - Browse (
browse_csdk_docsor resources) — when you know which category or file you need. Returns the full structured document, organized by the monorepo's own folder structure.
The search_csdk_docs tool also supports:
- Framework filter — restrict results to React, Vue, or Angular docs
- Code context (optional) — pass a code snippet and it extracts SDK imports, components, hooks, and factories to boost relevant results
Install
git clone https://github.com/pmorris24/csdk-docs-mcp.git
cd csdk-docs-mcp
npm install && npm run build
Connect to Claude Code
The repo includes .mcp.json — just open Claude Code in the repo directory and the server connects automatically.
Or add to any project's .mcp.json:
{
"mcpServers": {
"csdk-docs": {
"command": "node",
"args": ["/absolute/path/to/csdk-docs-mcp/dist/index.js"]
}
}
}
Connect to Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"csdk-docs": {
"command": "node",
"args": ["/absolute/path/to/csdk-docs-mcp/dist/index.js"]
}
}
}
Usage
Once connected, Claude will automatically use the tools when it needs SDK information. Just ask naturally:
"How do I query data with useExecuteQuery and display it in a ColumnChart?"
"What are the props for ChartWidget?"
"How do I embed a Fusion dashboard using DashboardById?"
"How do I customize chart colors with styleOptions?"
"How do I use onBeforeRender to modify Highcharts options directly?"
"How do I build a custom dashboard layout with useComposedDashboard?"
Complements the Sisense MCP Server
This works alongside the official sisense-mcp-server:
| Server | What it does |
|---|---|
| Sisense MCP | Queries your Sisense instance — get data sources, fields, build charts |
| This server | Searches SDK documentation — how to write the code to use that data |
Use both together: Sisense MCP tells Claude what data you have, this server tells Claude how to build the UI with the Compose SDK.
Tag handling
The server respects Sisense SDK documentation tags:
| Tag | Behavior |
|---|---|
@internal |
Completely excluded from search results and browse output |
@sisenseInternal |
Completely excluded from search results and browse output |
@beta / @alpha |
Included but flagged with a warning: "This API is marked as beta. It may change without notice." |
fusionEmbed |
Included normally (indicates Fusion embedding features) |
Updating docs
Docs are pulled directly from the official Sisense Compose SDK monorepo (docs-md/sdk/). When Sisense releases a new SDK version:
./scripts/update-docs.sh
npm run build
To pull from a specific branch:
./scripts/update-docs.sh --branch dev
Validation
Run the validation script after updating docs to verify completeness:
./scripts/validate-docs.sh
This checks:
- chunks.json exists and has expected chunk count
- All framework categories have doc files and INDEX.md
- Key files are present and non-empty
- No
@internal/@sisenseInternalcontent leaked into public docs - Framework parity (React, Vue, Angular have equal file counts)
- Build output is up to date
Supported frameworks
- React
- Vue
- Angular
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.