AI & data platforms | Modelled case study
Govern credentials used by autonomous AI agents
Which agent instance may call which tool, with what credential, data boundary, and expiration?
The modelled organisation
A recognisable problem reaches the operating agenda
This composite scenario follows the AI Platform · Identity Security · Application Owners functions. It is grounded in the cited problem context but does not identify a real customer.
Operating environment
An AI platform moves datasets, code, models, evaluations, logs, agent identities, credentials, and approvals through training and production systems.
What is at stake
A model or autonomous action is not trustworthy unless the platform can connect it to approved inputs, accountable identities, policy, and tamper-evident evidence.
Situation
Long-lived API keys and shared service accounts let an agent's prompt or plugin compromise become broad infrastructure access.
Event that forces action
Agent rollout, MCP integration, tool expansion, or credential misuse.
Concrete system boundary
Systems this case study puts in scope
The model is specific about the operational surfaces that must be discovered, changed, or independently checked.
agent identities
tool credentials
policy and approval service
session audit, revocation, and shutdown
Modelled case study walkthrough
How this organisation would use QNSI
The walkthrough connects the real-world problem to a bounded QNSI contribution and an independently reviewable result.
Recognise the operating condition
Long-lived API keys and shared service accounts let an agent's prompt or plugin compromise become broad infrastructure access.
Frame the decision the owners must make
Which agent instance may call which tool, with what credential, data boundary, and expiration?
Apply QNSI to the controlled boundary
Inventory agent and tool identities, keys, scopes, algorithms, owners, environments, and rotation events in QNSI.
Leave the team with a concrete result
An agent credential register with least-privilege scope, per-tool trust, session lifetime, revocation, and orphan detection.
Prove the result in the organisation's environment
The operator tests authorization, prompt-injection containment, secret isolation, human approval, audit completeness, and shutdown.
What useful success looks like
A decision artifact plus proof from the real environment
The model stops at a target result. It becomes an actual case study only when a customer produces and independently validates this evidence in production.
Decision artifact
An agent credential register with least-privilege scope, per-tool trust, session lifetime, revocation, and orphan detection.
Independent validation boundary
The operator tests authorization, prompt-injection containment, secret isolation, human approval, audit completeness, and shutdown.
Real-world problem grounding
Primary sources behind the model
These sources establish the external requirement, failure mode, or risk context used to model this case. They do not endorse HEOSSI or prove that QNSI completed the scenario.
Customer evidence status
This is modelled, not a customer claim
The organisation is a composite and the result is a target state. This page does not prove a deployment, customer outcome, certification, legal conclusion, regulator endorsement, or completed control.