Skip to main content
MCP Prompts are pre-built conversation templates that guide AI agents through complex multi-step workflows. When invoked, they inject structured instructions into the agent’s context window, ensuring consistent and thorough execution.

Available Prompts

Wallet Setup

Guided wallet creation and initial configuration.

Risk Assessment

Evaluate the safety of a pending on-chain transaction.

Security Audit

Comprehensive security review across all wallets.

Daily Report

Summarize daily operations and flag anomalies.

x402 Payment

Guided x402 payment flow with cost confirmation.

How Prompts Work

Unlike tools (which execute code) and resources (which return data), prompts inject structured instructions into the LLM’s context. This gives the agent:
  1. A clear task definition — what to analyze, in what order
  2. Expected output format — ensuring consistent, parseable results
  3. Safety guardrails — forcing the agent to review policies and constraints before acting
  4. Explainability — the agent produces a text-based reasoning trail before executing any action

Prompt Reference

PromptCategoryPurpose
wallet_setupSetupCreate and configure a new agent wallet
risk_assessmentAnalysisPre-transaction safety evaluation
security_auditAnalysisCross-wallet security posture review
daily_reportAnalysisOperational summary and anomaly detection
x402_paymentPaymentGuided cost discovery and payment flow