DevnorsAI/devnors-data-mcp

What it solves

This project provides a Model Context Protocol (MCP) server that allows AI assistants (like Claude Desktop, Cursor, and Codex) to directly access and query Devnors Data's extensive API suite. It bridges the gap between a general-purpose LLM and specialized data sources, enabling the assistant to perform real-world data retrieval tasks without the user needing to manually copy-paste information.

How it works

The server implements the MCP standard, allowing it to be integrated as either a remote HTTP server or a local stdio server. Once connected, the AI assistant can discover the available tools (capabilities) and their required parameters. The assistant then calls these tools to fetch data from Devnors Data APIs, such as legal documents, company registration details, and academic papers, using an API key for authentication.

Who it’s for

Users of MCP-compatible AI assistants who need access to professional data for research, legal analysis, company due diligence, and market research within their AI chat interface.

Highlights

• Broad Data Coverage: Access to legal cases, PRC law articles, enterprise registration, tax records, and academic research papers.

• Flexible Deployment: Supports both remote HTTP (no local installation needed) and local stdio (via Python package) configurations.

• $ ext{MCP}$ Standardized: Works with a wide range of AI tools including Codex, Claude Desktop, and Cursor.

• Self-Correcting Capabilities: Provides structured error codes (like insufficient_balance or rate_limited) that allow the AI agent to explain the issue and suggest the next steps to the user.