SteadyBuddy
SteadyBuddy is part of Steadybit Labs and is currently only available upon request. Learn more
SteadyBuddy is Steadybit's AI-powered assistant built into the platform that helps you design, run, and understand chaos experiments using natural language. Instead of clicking through the experiment editor, you can describe what you want to test, ask why an experiment failed, or let SteadyBuddy propose ready-to-run experiments tailored to your environment.
How SteadyBuddy Works
SteadyBuddy acts on your behalf and only sees what you are allowed to see. Every question, suggestion, and experiment draft is scoped to:
Your team — only the actions, environments, and targets available to your currently selected team are used as context.
Your permissions — SteadyBuddy reads the same data you can read in the UI/API. It cannot surface targets, environments, or runs you have no access to.
SteadyBuddy never performs destructive operations on its own. Today, it performs read-only operations: it reads your actions, environments, targets, advice, and experiment runs to answer questions and to draft experiments. It does not create, modify, or delete anything without an explicit, human-initiated action — for example, you clicking a button in the UI.
As the assistant gains the ability to take more actions on your behalf, we will add additional safeguards around potentially destructive operations.
Availability and Setup
How SteadyBuddy is enabled depends on whether you use the Steadybit SaaS platform or run Steadybit on-prem.
SaaS — Opt In
For SaaS customers, SteadyBuddy is off by default and must be explicitly enabled by a tenant administrator. This is an opt-in step because chat and suggestions are processed by an external model provider (see Data Processing and Privacy).
To enable SteadyBuddy, access it from the navigation or go to Settings → Data & Access. Once enabled, every user within your tenant can start using SteadyBuddy.
If you are not an administrator, ask one of your tenant admins to enable it for you.
On-Prem — Configure a Provider
For self-hosted (on-prem) installations, there is no separate opt-in: SteadyBuddy is available once you configure a model provider in the platform's environment variables and the AI capability is part of your license. On-prem deployments run against your own AI provider.
See configuration options / AI for the provider, model, and retention settings.
Data Processing and Privacy
SteadyBuddy only sends the context needed to answer your request. Depending on what you do, this can include: your typed messages, the names and metadata of actions and environments available to your team, target type and attribute summaries for the selected environment, reliability advice, and experiment run results. It never sends credentials or secrets.
SaaS: SteadyBuddy uses Anthropic models. Your data is not used to train any models — this is covered by our data privacy terms.
On-prem: processing happens through your own chosen AI provider. No data ever leaves your infrastructure — neither to a Steadybit-hosted service nor to any other AI provider.
Chat history is retained for a limited time (30 days by default), and you can delete any individual chat at any time. On-prem, retention periods are configurable.
What You Can Do
Get Experiment Suggestions
Pick an environment and SteadyBuddy analyzes the targets it contains to propose ready-to-run experiment ideas tailored to what it finds. Each suggestion comes with a short description and a More details view; from there you can Build experiment to open it in the editor.
Use this when you are getting started in a new environment, or need guidance on what to test next.
Experiment suggestions use the same flow described in the following section.
Create Experiments from Scratch via Chat
Describe the experiment you want in your own words — for example, "Test how my checkout service behaves when one of its pods is killed". SteadyBuddy asks clarifying questions when needed, then produces an experiment draft. You can:
Open Experiment Draft to review and edit it in the experiment editor, or
Run Experiment to execute it immediately.
The draft is scoped to your selected environment and uses only actions and targets your team can access.
The experiment draft must be saved manually before it appears in the Experiments section or can be run later.
Analyze an Experiment Run
Ask questions like "Why did my last experiment fail?". SteadyBuddy reads the relevant run data and explains what happened, grounding its answer in the actual execution results rather than guessing.
Work With Your Chat History
SteadyBuddy keeps a history of your past conversations in a collapsible sidebar on the left. Use the panel toggle to show or hide it, and New chat to start a fresh conversation at any time.
Your previous conversations appear under Recent chats. From there you can:
Open any past chat to continue where you left off.
Rename a chat to give it a more descriptive title — hover over an entry and open its ⋯ menu, then choose Rename.
Delete chat from the same ⋯ menu to remove a conversation you no longer need.
Search across your chats to quickly find a specific conversation by its title.
Your chat history is scoped to you and your currently selected team, and follows the same retention rules as the rest of SteadyBuddy.
Usage Limits
AI usage is subject to a budget tied to your plan. When the budget for the current period is reached, the chat and suggestions are not available anymore. Please contact us to enable AI features again.
Last updated
Was this helpful?