--- name: al-341-give-your-published-app-an-ai-on-call-engineer-let-replit description: "Most people who build an app alone find out it broke when a user writes 'doesn't work for me'. Replit launched Monitoring Agent on Oct 22, 2026 as an autonomous on-call engineer for apps published on Replit. Its automatic checks watch three signals (application error logs, HTTP errors and request latency); with Auto-fix on, it investigates an alert using production metrics, runtime logs and your code, attempts and validates a fix, and emails you a summary. Auto-republish, which lets it put the fix live on its own, is a separate, experimental setting that requires Auto-fix, and Replit warns some changes such as database migrations and index operations may be risky. Alerts, Auto-fix and Auto-republish are set per app in Publishing settings, runs may incur Agent usage charges, and you can also just ask Agent about your production logs and metrics. The prompt below is a safe first investigation you can run any time, and a review step before you publish a fix. Use when the user wants help…" --- # Give your published app an AI on-call engineer: let Replit's Monitoring Agent investigate errors and slowdowns, draft and test a fix, and keep publishing in your hands until you trust it Most people who build an app alone find out it broke when a user writes 'doesn't work for me'. Replit launched Monitoring Agent on Oct 22, 2026 as an autonomous on-call engineer for apps published on Replit. Its automatic checks watch three signals (application error logs, HTTP errors and request latency); with Auto-fix on, it investigates an alert using production metrics, runtime logs and your code, attempts and validates a fix, and emails you a summary. Auto-republish, which lets it put the fix live on its own, is a separate, experimental setting that requires Auto-fix, and Replit warns some changes such as database migrations and index operations may be risky. Alerts, Auto-fix and Auto-republish are set per app in Publishing settings, runs may incur Agent usage charges, and you can also just ask Agent about your production logs and metrics. The prompt below is a safe first investigation you can run any time, and a review step before you publish a fix. ## Inputs to collect first Ask the user for anything below that they haven't already given you. Don't guess missing facts. - App name - What it does and who uses it - What users reported or what I noticed, e.g. 500 errors / slow… - Approximate time and time zone - Endpoint or page, if known ## Instructions Do the task below with the user's inputs in place of each [bracket]. You are my on-call engineer for my published app, [app name], which [what it does and who uses it]. Something looks wrong: [what users reported or what I noticed, e.g. 500 errors / slow checkout page / failed deploy], starting around [approximate time and time zone], affecting [endpoint or page, if known]. 1) Investigate first, change nothing: check runtime logs, error counts, HTTP 4xx/5xx responses and latency (p50 and p99) for that time window, plus CPU and memory since my last publish. 2) Tell me in plain English: what broke, when it started, how many users were likely affected, and your best guess at the cause, with the log lines or numbers that support it. Say clearly if the cause is outside my app (e.g. a third-party service is down) and what I can do about it. 3) If a code fix is needed, propose the smallest change, explain it in one paragraph, and test it. Do not touch database migrations, indexes or stored data without asking me. 4) Don't publish anything. Give me a short summary I can review: what changed, how you tested it, what could go wrong, and how to roll it back. 5) Suggest one alert or check that would have caught this sooner. ## Safety checks (share these with the user) - Read and edit the draft before it reaches a customer or goes public. If you later automate this prompt, start with "draft and notify me", not "send automatically". - Check names, dates, numbers and any claim against your own source material before you use the answer. AI can misread or invent details. - Say clearly which parts need a person to review or approve before anything is sent, paid, posted or filed. ## About this skill - Works with: Replit Agent, Replit Monitoring Agent - Based on this real-world example: https://replit.com/blog/replit-monitoring-agent - Free prompt from AgentPlaybook Real Estate: https://agentic-library.netlify.app/prompt/al-341-give-your-published-app-an-ai-on-call-engineer-let-replit-s/ - To install in an agent that supports Agent Skills, save this file as SKILL.md in a folder with the same name as `name` above (for example in your skills directory). Install steps for Claude.ai, Claude Code, Cursor and Codex: https://agentic-library.netlify.app/skills/