Cloud Security Platform · Developer Portal, API & MCP Server
Ask about your estate in plain English
Ayurak Intelligence answers questions about your own estate with its sources cited, and any change it proposes waits for a person to approve it.
The problem
I inherited an alert queue, and getting context on one finding means opening several screens. An assistant that can change infrastructure worries me unless there is a hard stop between a suggestion and an action. If an automated change fails halfway, I need it on the record.
How Aribot handles it
- 1
Ask a question
Type a question in plain English, such as which threats sit on your latest diagram.
- 2
Get a cited answer
The answer names the diagram and account it was read from, so you can check it.
- 3
Review a proposed fix
Suggested changes appear as proposed actions at pending approval, with nothing changed.
- 4
Approve or roll back
Approved actions run, and a change that fails mid-apply is rolled back and kept on the list.
What you get
- Answers drawn from your own diagrams and accounts
- The source cited with every answer
- Proposed changes held at pending approval
- Failed changes rolled back and recorded
Related use cases
Let AI assistants call Aribot through one governed gateway
Point any MCP client at the Aribot server, and its calls pass through the same gateway and permission checks as every other agent action, with write calls metered.
AI-drafted fixes that wait for your approval
Aribot drafts a fix for each finding with the model that wrote it on record, and nothing changes until a person approves the dry run.