Web3 Research MCP

You can run the Web3 Research MCP as a local or remote service that orchestrates token research using multiple sources, stores results for reference, and tracks progress across sections. It helps you build comprehensive token reports without leaving your development environment by simplifying setup, usage, and automation with MCP clients.

How to use

You will interact with the MCP server through an MCP client. Start by ensuring you have an MCP client installed, then start the Web3 Research MCP server via the commands provided. Use the client to create structured research plans, run searches across multiple data sources, fetch content, and update the progress status of each research section. All resources generated during a research session are saved automatically for reference, and you can list or fetch them later to build a complete report.

How to install

# Prerequisites
node -v
# You should have Node.js v16 or higher

# Install the MCP client tooling for the Web3 Research MCP via Smithery
npx -y @smithery/cli install web3-research-mcp --client claude

Note: You will also add an MCP server entry to your client configuration to enable integration with Claude Desktop or Cursor as shown in the configuration steps below.

## Configuration and usage notes

Configure your MCP client to connect to the Web3 Research MCP using the provided commands. This enables you to run the research workflow from your preferred environment and track progress across sections.

{ "mcpServers": { "web3_research": { "command": "npx", "args": ["-y", "web3-research-mcp@latest"] } } }

## Operational workflow

1) Create a structured research plan for a token. 2) Perform searches across sources like CoinGecko, CoinMarketCap, and DeFiLlama. 3) Save results as resources for reference. 4) Update the status of each research section as you progress. 5) Generate a comprehensive report that covers technical fundamentals, market data, social sentiment, and other relevant insights.

## Security and maintenance notes

Keep your MCP client and server dependencies up to date. Be mindful of rate limits and access restrictions on certain websites. If you encounter 403 errors during content fetching, you can rely on stored resources or adjust search scopes to avoid restricted sources.

## Troubleshooting tips

If the MCP server does not respond, verify that the client configuration points to the correct command and arguments, and ensure Node.js is available in your environment. Check for network issues or blocked endpoints that could prevent searches from completing.

## Notes on usage with Claude Desktop and Cursor

To use with Claude Desktop, add the MCP server configuration to your Claude Desktop config and restart Claude Desktop. The example uses npx to invoke the latest version of the MCP.

To use with Cursor, add the MCP server configuration to your Cursor MCP settings and restart Cursor.

## Available tools

### create-research-plan

Creates a structured research plan for a token using the token name and ticker to outline research sections.

### search

Performs a web search and returns results across web, news, images, or videos.

### research-with-keywords

Searches for a token with specific keywords and saves the results for later reference.

### update-status

Updates the status of a research section (planned, in\_progress, completed).

### fetch-content

Fetches content from a URL and saves it as a resource in the repository.

### list-resources

Lists all saved resources available for reference during the research.

### search-source

Searches for token information from a specific source such as CoinGecko or DeFiLlama.