AI in practice
Using AI Sidekick and your own AI agent day to day. Your favourite AI platform, and concrete examples of what gets faster.
There are two ways to bring AI into how you run this: AI Sidekick, built into the platform, or your own agent connecting over MCP from whichever AI product you already use.
Using it from your own AI platform
Bringing your own agent works from any MCP-compatible client (Claude, Copilot, Cursor, or a self-hosted model), using your own tokens rather than the platform's. See Connect your AI agent for the actual setup: the endpoint, authentication and what an agent is and isn't allowed to do on its own.
What it speeds up
A few concrete examples of the difference this makes in practice:
- Triaging a batch of findings in one request, instead of opening each one individually to accept, suppress or resolve it.
- Generating a one-page executive summary from a single prompt, instead of assembling one by hand from several dashboard views.
- Asking a specific, plainly-worded question about your posture, such as "what's affected by this CVE" or "what changed since last week", and getting a direct answer instead of piecing it together across Exposure, Risk and Compliance.
None of this needs the platform's own AI credits. See the AI-credits answer for how that's metered separately.
API keys
Create and manage personal and agent API keys. Where each screen lives, key types, expiry, and how revoking interacts with your organisation's agent-access policy.
How it works
How Dashboard, AI Sidekick, Assets, Exposure, penetration tests and assessments, Governance, Risk, Compliance, Actions, Activity and Preferences connect into one continuous loop. Read once to see how the platform fits together.