Skip to main content

Prompt Cookbook

These prompts are written for natural-language MCP clients. They assume the client will map your request to the correct Rug Radar tool.

Token Research​

  • Analyze <token address> and give me a plain-English verdict.
  • Research <token name>. Focus on holder concentration and contract risk.
  • Analyze <token> and tell me the three biggest reasons to avoid it.
  • Analyze <token> and explain whether the risk is structural or just early-stage volatility.

Fast Safety​

  • Scan red flags for <token address>. Keep it short.
  • Is <token> safe right now? Show only critical flags.
  • Give me a go/no-go scan for <token> with only triggered flags.

Wallet Analysis​

  • Analyze this wallet: <wallet>. Is it sniper, whale, bot, insider, or retail?
  • Check this wallet's behavior and tell me if I should mirror it.
  • Analyze this wallet and explain what drove the classification.

Token Comparison​

  • Compare <token A> vs <token B> and pick the healthier one.
  • Which is safer between <token A> and <token B>, and why?
  • Compare <A> vs <B> focusing on liquidity, concentration, and contract safety.

Monitoring​

  • Watch <token address> with 20% liquidity-drop threshold.
  • What changed since last check for <token address>?
  • Watch <token> with 10 point risk score increase and authority change alerts.
  • Compare the current state of <token> to the last stored snapshot and highlight only material changes.

Prompting Tips​

  • say whether you want a short verdict or a full report
  • use token addresses when precision matters
  • ask for plain English if you are sharing the output with non-technical users
  • ask for only red flags when speed matters more than completeness