About solana-mcp
solana-mcp (Rigel) is a read-only MCP server for agents and other MCP clients inspecting Solana and pump.fun data. It combines wallet, token, deployer, risk, and live market checks without handling keys or signing transactions.
Key features
- 01Reads SOL balances and SPL token holdings
- 02Inspects token metadata and bonding curves
- 03Assesses authority and holder-concentration risks
- 04Checks first buyers and deployer history
- 05Returns live market data and full audits
Use cases
- 01Review a wallet’s SOL and token holdings
- 02Inspect a pump.fun token before further review
- 03Check token risk signals and buyer patterns
- 04Get a combined on-chain and market snapshot
MCP support
The server exposes seven read-only Solana tools to MCP clients over stdio, including wallet, token, risk, deployer, market, and full-audit queries.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y @rigel-protocol/solana-mcpSOLANA_RPC_URLSolana RPC endpoint to read from. Public mainnet by default; a Helius/Triton key is recommended.Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
7Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| walletAddress | string | Yes | Base58 Solana wallet address to inspect |
Details
From the project README✦ Rigel — Solana MCP server
Give your AI agent eyes on Solana. Rigel is an open-source MCP server that connects Claude Desktop, Claude Code, Cursor, and any stdio MCP client to the Solana blockchain and the pump.fun ecosystem.
Read-only by design. No private keys, no signing, no custody — the only configuration is which RPC to read from.
Tools
| Tool | Input | Returns |
|---|---|---|
rigel_get_wallet_balance | walletAddress | SOL balance + major SPL token holdings |
rigel_inspect_token_data | tokenAddress | Metadata, supply, bonding-curve SOL, curve completion % |
rigel_analyze_rug_risk | tokenAddress | Mint/freeze authority status, top-holder concentration, risk rating |
rigel_check_first_buyers | tokenAddress | Bundle detection: fresh-wallet holders, shared funding wallets, supply concentration |
rigel_deployer_history | tokenAddress | Deployer rap sheet: who launched it, wallet age, activity, remaining bag |
rigel_market_snapshot | tokenAddress | Live price, market cap, liquidity, volume, buys/sells flow, momentum read |
rigel_full_audit | tokenAddress | The whole battery in one call, with a combined verdict and top reasons |
Quick start (Claude Desktop)
Paste into claude_desktop_config.json and restart Claude:
{
"mcpServers": {
"rigel-solana": {
"command": "npx",
"args": ["-y", "@rigel-protocol/solana-mcp"],
"env": { "SOLANA_RPC_URL": "https://api.mainnet-beta.solana.com" }
}
}
}
Then ask your agent things like:
- "What's in this wallet?
FC5i…" - "Inspect this pump.fun token:
7xKq…" - "Is this token a rug?"
Build from source
npm install
npm run build
node dist/index.js # speaks MCP over stdio
Configuration
| Env var | Default | Purpose |
|---|---|---|
SOLANA_RPC_URL | https://api.mainnet-beta.solana.com | Any Solana RPC — public, Helius, Triton, or your own node |
The public mainnet RPC rate-limits aggressively; a free Helius key is recommended for heavy use.
Security posture
- No private keys, seed phrases, or signing code anywhere in this codebase.
- Local stdio transport — queries go from your machine to your RPC, nothing in between.
- Logs go to stderr; stdout is reserved for MCP protocol traffic.
- Zod-validated inputs; friendly errors for invalid public keys and RPC failures.
Disclaimer
Rigel reports on-chain facts. The rug-risk rating is a heuristic over public data (authority status, holder concentration) — it cannot detect every risk, and nothing it outputs is financial advice.
License
MIT
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.