AppSignal Launch Week 2026 Recap

That's a wrap on Launch Week 2026. In case you missed a day, or you want the short version, here is everything we shipped.

Day 1: Render Metrics Stream Support

Render makes deploys easy. When something goes wrong, the signal is often scattered across HTTP latency, Postgres connections, and host CPU, not a single 500 page.

AppSignal now supports Render's Metrics Stream. Set it up once per Render workspace, and Render forwards OpenTelemetry metrics to our Collector. Within minutes you get host metrics for every Render service, plus automated dashboards for HTTP, Postgres, and Key Value, without a config file per service.

Watch the Render Metrics Stream announcementWatch the Render Metrics Stream announcement video

Day 2: Hosted MCP + CLI

We shipped two ways to bring production data into the tools you already use.

Hosted MCP: no need to run a Docker container to connect your AI agent. Point Claude Code, Cursor, Zed, or any MCP-compatible client at https://appsignal.com/api/mcp, sign in, and you are done. Errors, performance, logs, metrics, and dashboards are queryable in plain English from your editor, for any app or environment you monitor.

A public CLI: a single Rust binary. Install with Homebrew or a script, then tail logs, search incidents, and manage triggers from your terminal. It ships with an agent skill, so Claude, Codex, or OpenCode can call AppSignal without glue code.

Watch the Hosted MCP and CLI announcementWatch the Hosted MCP and CLI announcement video

Day 3: Logging

SSHing in to tail a file only gets you so far. This week we gave our query language more range: native AND, OR, and nested grouping, dot notation into JSON, and comparison operators such as duration_ms>2000. That means queries like:

severity=error environment=production user.id=12345

Beyond search, you can turn any log line into a metric (count, gauge, or measurement) or a trigger from the details panel. Set it up once and it shows up on dashboards and in anomaly detection alongside everything else.

Watch the Logging announcementWatch the Logging announcement video

Day 4: Process Monitoring

Cron jobs and background workers fail quietly. No 500 page, no warning, nothing, until something downstream breaks and a customer finds it first.

Process Monitoring covers cron jobs and heartbeat processes such as Sidekiq or queue consumers. Wire it up with a helper in your AppSignal integration, or wrap any command with appsignal-wrap, no code changes required. Every run gets occurrence-level detail, linked to the error, slow query, or log lines from that run.

Watch the Process Monitoring announcementWatch the Process Monitoring announcement video

Day 5: Dark Mode

We closed the week with something for your eyes: dark mode for the AppSignal UI. Light, dark, or system. Set it once on your account and it follows you across devices. Useful for late deploys and later on-call shifts.

Read the Dark Mode announcement

Everything Is Live

Five days, five features. Thanks for following along, whether that was live each morning or in one sitting now.

Everything we shipped this week is live today, on every plan, including our free plan. If you want to try it:

• Start a 30-day free trial, no credit card required
• Already a customer? Update your integrations and check the docs for setup guides
• Questions or feedback? Email support@appsignal.com. Real engineers answer.

See you at the next one.