Hermes Agent + GoldRush

Set up Hermes Agent with GoldRush

1. 1Install Hermes Agent

   curl -fsSL https://raw.githubusercontent.com/NousResearch/hermes-agent/main/scripts/install.sh | bash

2. 2Enable MCP support

   Install the MCP extras if your current Hermes environment does not already include them.

   pip install "hermes-agent[mcp]"

3. 3Run the Hermes setup wizard

   Configure your preferred inference provider before you connect external tools.

   hermes setup

4. 4Add GoldRush MCP server to ~/.hermes/config.yaml

   mcp_servers:
     goldrush:
       command: "npx"
       args: ["-y", "@covalenthq/goldrush-mcp-server"]
       env:
         GOLDRUSH_API_KEY: "YOUR_API_KEY"

5. 5Start Hermes and assign a blockchain task

   Ask: "Use GoldRush to monitor this wallet, summarize new transfers every hour, and remember the reporting format I prefer"

   hermes

Things to ask Hermes Agent

Once GoldRush is connected, ask in plain words. Hermes Agent picks the calls.

• “Create a reusable Hermes skill for wallet portfolio analysis with GoldRush”
• “Spawn subagents to compare token balances across Ethereum, Base, and Arbitrum”
• “Set up a recurring wallet monitor and notify me when transfers exceed a threshold”
• “Use GoldRush to build a daily markdown report for this treasury wallet”
• “Write and execute code that classifies this wallet's recent transactions”
• “Remember my preferred chain list and reporting layout for future GoldRush tasks”
• “Investigate suspicious token approvals for this address and summarize the risk”
• “Run a chain-by-chain analysis and aggregate the results into one report”

Common workflows

• Persistent wallet monitoring

  1. Configure GoldRush once in ~/.hermes/config.yaml
  2. Ask Hermes to watch a wallet and remember your preferred alert format
  3. Let Hermes call GoldRush on a schedule and store the workflow as a skill
  4. Reuse the same long-running agent instead of rebuilding context each session

• Parallel multi-chain analysis

  1. Give Hermes a wallet or protocol to analyze
  2. Ask it to spawn subagents for different chains or data domains
  3. Each subagent queries GoldRush independently and reports back
  4. Hermes merges the findings into a single summary or action plan

• Teach Hermes a GoldRush skill

  1. Solve a real blockchain task end-to-end with Hermes once
  2. Ask it to package the winning approach into a reusable skill
  3. Store the skill so future GoldRush tasks start from working procedure, not scratch
  4. Share that skill across environments if your team reuses the same workflow

• Server-side blockchain automation

  1. Deploy Hermes on a VPS or server environment
  2. Connect GoldRush through MCP and set the required API key
  3. Run long-lived treasury, whale-tracking, or compliance workflows
  4. Use Hermes memory to keep user preferences and reporting conventions stable

npx skills add covalenthq/goldrush-agent-skills

• Foundational API

  REST endpoints for historical blockchain data. Balances, transactions, NFTs, token prices, security checks, cross-chain queries across 100+ networks.

• x402

  Pay-per-request access without API keys. Agents pay with stablecoins using the HTTP 402 protocol. No accounts, no rate limits, no key management.

• CLI

  Terminal-first blockchain data with native MCP support. 17 commands that work for both humans and AI agents.

More agents

• Claude Code

  Anthropic's agentic coding tool that lives in your terminal and understands your codebase end-to-end.

• Codex

  OpenAI's cloud-native coding agent that runs tasks in a sandboxed environment with full autonomy.

• Gemini CLI

  Google's open-source terminal AI agent powered by Gemini models with built-in MCP support and a file-based skill system.