Blueprintd is a UK software company. We build redphonr, which lets AI agents ask a person before risky actions, and CronEcho, which watches your cron jobs and agents so failures never go unnoticed.
Human-in-the-loop for AI agents
Human approvals for AI agents, over MCP
redphonr pauses an AI agent before a risky action, asks a person on Slack, SMS, WhatsApp, Telegram, Discord, email or a phone call, and resumes the agent with their answer.
Cron job and AI agent monitoring
Monitoring for cron jobs and AI agents
CronEcho monitors cron jobs, background workflows and AI agents. It alerts you when a job misses its check-in or an agent breaks a budget or timeout, and escalates until someone acknowledges.
AI agents now send email, move money and change production systems. Most of the time they get it right. The failures that hurt are rare, quiet and expensive: a loop that burns through a budget overnight, a bulk delete with the wrong filter, a job that silently stops running.
We build the two things every team running agents ends up needing: a way for the agent to stop and ask a person, and a way to know when something it was supposed to do did not happen.
Step-by-step: connect the redphonr MCP server to Claude Code, Claude Desktop or Cursor, and make your agent ask you on Slack or SMS before risky actions.
We compared 10 ways to add human approval to AI agents: managed services, workflow tools and framework features. Channels, MCP support and when to use each.
From MCP-native escalation to PagerDuty, Twilio and Slack webhooks: 10 ways an AI agent can reach a person, ranked by how well they get an answer back.
Tell us what you are building, or what is stopping you from trusting an agent with more. We read every message.
Get in touch