dfns-mcp
Provides AI agents with access to DFNS documentation, API reference, and SDK code through full-text search and type intelligence.
README
DFNS MCP Server
A Model Context Protocol (MCP) server that provides AI agents with access to the DFNS documentation, API reference, and SDK code.
Features
- Full-Text Search: Instantly search across all DFNS guides, API docs, and SDK source code
- TypeScript Type Intelligence: Search and retrieve SDK type definitions with import paths
- Auto-Updating Docs: Documentation is fetched from docs.dfns.co on first run and cached locally
- Context Initialization: Dedicated
inittool to enforce documentation-first behavior - Smart Context: Retrieve specific documents, code examples, and API endpoint definitions
- SDK Intelligence: Mapped knowledge of supported blockchains and their SDK packages
Requirements
This MCP server requires Bun - a fast JavaScript runtime.
Linux/macOS:
curl -fsSL https://bun.sh/install | bash
Windows (PowerShell):
powershell -c "irm bun.sh/install.ps1 | iex"
Quick Start
Claude Code (Recommended)
claude mcp add -s user dfns-mcp -- bunx dfns-mcp@latest
The -s user flag installs to the user so it's available in all your projects.
Cursor
Add to your Cursor MCP settings (.cursor/mcp.json):
{
"mcpServers": {
"dfns-mcp": {
"command": "bunx",
"args": ["dfns-mcp@latest"]
}
}
}
Codex
codex mcp add dfns-mcp -- bunx dfns-mcp@latest
Gemini CLI
Add to your Gemini CLI config (~/.gemini/settings.json):
{
"mcpServers": {
"dfns-mcp": {
"command": "bunx",
"args": ["dfns-mcp@latest"]
}
}
}
Global Installation
If you prefer to install globally instead of using bunx:
bun install -g dfns-mcp
# Then reference as just "dfns-mcp" in your MCP config
Agent Instructions (Recommended)
Add this to your global ~/.claude/CLAUDE.md (or ~/AGENTS.md for other agents) to ensure your AI always uses up-to-date DFNS documentation:
## DFNS Development
When working with DFNS (wallet infrastructure, key management, blockchain integrations):
1. **Always** call `mcp__dfns-mcp__init` at the start of any DFNS-related task
2. **Never** rely on training data for DFNS APIs - always use `search_docs` and `get_doc` to verify
3. Use `search_types` and `get_type` to get accurate TypeScript type definitions
4. DFNS APIs and SDKs change frequently - the MCP server has the latest documentation
This prevents agents from hallucinating outdated API signatures or SDK patterns.
How It Works
On first run, the server automatically downloads the latest DFNS documentation from docs.dfns.co and SDK types from GitHub, caching them locally in ~/.cache/dfns-mcp/. The cache is refreshed every 24 hours automatically, or you can force an update using the update_docs tool.
Available Tools
Documentation Tools
| Tool | Description |
|---|---|
init |
CALL THIS FIRST. Initializes the session and mandates strict documentation adherence |
search_docs |
Search all documentation and SDK files |
get_doc |
Retrieve the full content of a specific document |
list_docs |
List available documents by category |
get_code_examples |
Extract code snippets for a specific topic |
browse_api_structure |
View hierarchical API endpoint structure |
get_api_endpoint |
Get details for a specific API endpoint (e.g., POST /wallets) |
get_blockchain_info |
Get SDK package info for a specific chain (e.g., Solana) |
list_blockchains |
List all supported blockchains with their SDK packages |
TypeScript Type Tools
| Tool | Description |
|---|---|
search_types |
Search SDK types by name (e.g., "Wallet", "Signer", "Transaction") |
get_type |
Get full type definition, import path, and usage |
list_types |
List all types by category |
Cache Management
| Tool | Description |
|---|---|
update_docs |
Force update the documentation cache from docs.dfns.co and GitHub |
cache_info |
Get information about the cache (location, last update, etc.) |
Resources
The server also exposes these quick-reference resources:
| Resource URI | Description |
|---|---|
dfns://quickref/authentication |
Authentication patterns and code samples |
dfns://quickref/sdk-setup |
Package installation and setup guide |
dfns://quickref/networks |
All supported blockchain networks |
Best Practices for Agents
When using this server, agents should follow this workflow:
- Initialize: Call
initimmediately to establish the "Documentation First" protocol - Search: Use
search_docsto find relevant information before answering ANY question - Read: Use
get_docto read the actual source material - Answer: Formulate responses based only on the retrieved context
Development
# Clone the repo
git clone https://github.com/jmaddington/dfns-mcp.git
cd dfns-mcp
# Install dependencies
bun install
# Run in development mode with auto-reload
bun run dev
# Test with MCP Inspector
bun run inspect
License
MIT
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.