Home · Offerings · AI Launch Check

Offerings

AI Launch Check One-time · start here

Before an AI feature goes live.

Know what your new AI feature can read, what it can reach, and what could go wrong — before your customers or their security team find out.

Who it's for

Teams about to ship AI.

You're about to launch an assistant, an agent or an AI search. Get the risks written down while they're cheap to fix. Everything carries forward into an AI Red Team.

What's included

Four checks before go-live.

  • AI inventory (AI-BOM)Every model, agent, MCP server, tool and data source the feature touches.
  • Threat model and data flowsWhere untrusted content can enter, and what the model can reach once it does.
  • Agent permission reviewWhat each agent and tool may do, and what stops it doing more.
  • Go-live risk reportWhat could go wrong, written as actions, ordered by what to fix first.

What you receive

Written down, ready to act on.

  • The AI-BOM for the feature
  • A threat model with its data-flow map
  • The permission review findings
  • A go-live risk report your engineers can work through
  • Every risk mapped to the OWASP Top 10 for LLM and Agentic Applications

Built for

The systems we check.

In our clients' words

Scoped, evidenced, actionable.

Korelex tested Nelogi’s AI workflows and access boundaries within an agreed scope. Their report provided clear, evidence-backed findings, actionable reproduction steps and prioritised remediation recommendations, helping our engineers plan the next steps.
Natesan KumaresanCISSP, CCSP, CISM, CISA · Founder, Nelogi & SydanticIQ

Next step

Check it before it ships.

Tell us what you're launching. We'll scope the check around it.

Pricing shared on request.