dfns-mcp
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@dfns-mcpsearch docs for creating a wallet"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
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 behaviorSmart Context: Retrieve specific documents, code examples, and API endpoint definitions
SDK Intelligence: Mapped knowledge of supported blockchains and their SDK packages
Related MCP server: Multi-Source Documentation MCP Server
Requirements
This MCP server requires Bun - a fast JavaScript runtime.
Linux/macOS:
curl -fsSL https://bun.sh/install | bashWindows (PowerShell):
powershell -c "irm bun.sh/install.ps1 | iex"Quick Start
Claude Code (Recommended)
claude mcp add -s user dfns-mcp -- bunx dfns-mcp@latestThe -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@latestGemini 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 configAgent 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 documentationThis 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 |
| CALL THIS FIRST. Initializes the session and mandates strict documentation adherence |
| Search all documentation and SDK files |
| Retrieve the full content of a specific document |
| List available documents by category |
| Extract code snippets for a specific topic |
| View hierarchical API endpoint structure |
| Get details for a specific API endpoint (e.g., |
| Get SDK package info for a specific chain (e.g., |
| List all supported blockchains with their SDK packages |
TypeScript Type Tools
Tool | Description |
| Search SDK types by name (e.g., "Wallet", "Signer", "Transaction") |
| Get full type definition, import path, and usage |
| List all types by category |
Cache Management
Tool | Description |
| Force update the documentation cache from docs.dfns.co and GitHub |
| Get information about the cache (location, last update, etc.) |
Resources
The server also exposes these quick-reference resources:
Resource URI | Description |
| Authentication patterns and code samples |
| Package installation and setup guide |
| 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" protocolSearch: Use
search_docsto find relevant information before answering ANY questionRead: Use
get_docto read the actual source materialAnswer: 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 inspectLicense
MIT
This server cannot be deployed
Maintenance
Related MCP Connectors
Versioned documentation registry and semantic search for AI tools and coding assistants.
Search @imqueue docs and scaffold typed services & clients from your AI coding agent.
Provide your AI coding tools with token-efficient access to up-to-date technical documentation for…
Shared knowledge base for AI agents. Semantic search across agents, no setup required — just a URL.
Related MCP Servers
- -licenseNot gradedqualityNot gradedmaintenanceEnables AI agents to intelligently search and retrieve Roblox documentation through semantic search and vector embeddings, providing natural language access to complete Roblox Creator Documentation.-
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to search and query documentation from multiple sources including Voiceflow and Claude Code, with full-text search, code examples retrieval, and step-by-step tutorials access.1MIT
- FlicenseNot gradedqualityDmaintenanceEnables AI agents to search and query documentation from git repositories using hybrid search and structured metadata queries.1-
- AlicenseAqualityDmaintenanceProvides access to Agno framework documentation for AI agents, enabling search and retrieval of SDK references, API endpoints, code examples, and integration guides through MCP-compatible tools.73MIT