Turn production investigations
into reusable runbooks.
Build investigation flows once. Run them during incidents with AI explaining the evidence step by step. Stop re-deriving the same diagnosis at 2am.
Free during public beta · macOS · Bring your own AI provider
Runs on your machine Bring your own AI
Live integrations · see what's coming next →
Who this is for
Built for teams with
repeated production investigations.
For teams where production diagnosis is repeated, distributed, and risky enough that ad-hoc SSH plus Claude is no longer acceptable. If you're diagnosing the same kind of incident more than once a month and the playbook lives in someone's head, you're in the right place.
For
DevOps agencies and MSPs
Many client environments. One diagnostic playbook. Junior technicians stop escalating to senior infra people.
For
SaaS with sprawling infra
10 to 100 services, queues, workers, cron jobs, and tenant-specific environments. One engineer can't hold it all.
For
Support engineering
Customer says the import is stuck. Inspect production safely without giving everyone shell access or escalating to backend.
For
Regulated teams
SOC 2, HIPAA, PCI. Every command, output, and AI response captured for the audit. Human review for risky actions.
The core object
The Diagnostic Runbook.
One mental model for every incident.
A repeatable sequence of SSH commands, HTTP checks, log queries, and AI interpretation steps that tells your team what broke and why. You build it once. The AI reads the output.
Real commands, real servers
SSH into the boxes you already own. Run the same checks you'd run by hand — only this time they're chained, repeatable, and shareable.
AI as a runbook step
Drop an AI step in anywhere. It reads prior outputs, summarises logs, and explains what the evidence suggests. You stay in the loop.
Evidence over hunches
Every command, every response, every AI explanation is captured. Hand off to a teammate, post-mortem on Monday, audit at any time.
See it in action
Watch a runbook walk through an incident.
Two minutes. One real-shape diagnostic runbook, run from SuperTerminal against a sample stack. Click play.
From alert to root cause
in minutes, not hours
Build a runbook once. Point it at an incident. The AI reads the output and tells you what broke.
Explore SuperTerminal
AI interprets.
You control execution.
Diagnostic runbooks are deterministic. You define the shell commands, HTTP requests, and AI steps up front. The AI reads the results, explains what it sees, and suggests what to check next.
You define the runbook
Commands, requests, and prompts are written by you, not invented on the fly.
No arbitrary commands
AI runbook steps can inspect prior results, but they do not execute shell or HTTP actions on their own.
Every step logged
Execution snapshots, outputs, and AI responses are saved so you can review what happened.
Runs on your machine
SuperTerminal is a local desktop app, and you choose which AI provider to use.
What the AI does here
- Interpret command output
- Summarise logs and stack traces
- Explain why a step produced what it did
- Suggest what to check next within the runbook
Shared runbooks.
Background investigations.
When an alert matches a runbook, BitSentry Team investigates automatically and queues an evidence-backed hypothesis. Your team sees the work — not just the result. Human review, audit trail, and shared library included.
See BitSentry TeamPick your starting point.
Run runbooks alone on your Mac, or roll them out across your team with a guided pilot.