AI Agent Made a Refund Without Approval
What to do right now and how to put a human and firewall in front of refunds so it cannot happen again.
Read →Agent Essays
Short, opinionated pieces about what it means for an agent to have identity, reputation, escrow, and the ability to pay for its own usage.
InfoSec Rejected Our AI Agent Deployment
Turn a security rejection into a concrete architecture with approval gates, encryption, and audit trails.
Read →LangGraph Production Human Approval Before Tool Execution
Pause LangGraph before dangerous tools and resume only after an operator approves via LetsPing.
Read →Autonomous Customer Service Bot on a Production Database
Pattern for support agents that read and write real customer data safely with LetsPing in the loop.
Read →Why Agents Need Credentials, Not Just API Keys
Give your agent a stable identity and credentials instead of sharing a human API key.
Read →Designing Agent Credit Bureaus: Trust Scores for Agents
How to turn agent behavior into a signed trust score other services can use.
Read →Agent-to-Agent Escrow and Handoffs
How agents make commitments to each other and leave a verifiable chain of custody.
Read →How Agents Pay for Themselves Without Your Stripe Keys
Let agents resolve 402s and fund usage while you keep control of billing.
Read →Add Human-in-the-Loop to OpenClaw in 5 Minutes
OpenClaw human approval and HITL. Pair your terminal, add the LetsPing skill, and approve high-risk actions from the PWA.
Read →