How can I chat with my Azure architecture?

Quick answer

Open a Cloudeval project and ask questions against its diagrams, resources, reports, and source details.

The answer can use project context instead of giving generic cloud advice.

Ask AI about a project Read MCP setup

What you can expect

A practical path from source input to diagram, report context, docs, exports, and the first AI question to ask.

Start here

Use this to

Ask architecture questions that are based on your Cloudeval project.

Best source to use

Ask one specific question about a resource, report, dependency, exposure path, or cost concern.

What you need

A Cloudeval project with imported or synced Azure context.
At least one diagram, report, or resource graph to ground the answer.
A specific question, not a broad prompt.

What you get

An answer tied to project context.
Evidence, citations, or report references where available.
A next step when evidence is missing.

Fastest path to a useful result

1
Open a Cloudeval project.
2
Run or review a report.
3
Ask one focused question about a resource, risk, dependency, or cost.
4
Check the evidence before sharing the answer.

Cloudeval question prompt for asking about cloud resources

Helpful docs

Report concepts Diagram concepts MCP setup Sharing and collaboration

When this works best

Works best after a project has an imported diagram, report, or live Azure sync.

If data is missing, ask what context is unavailable before acting.

Decision guide

Check	Cloudeval path	Note
Question style	Ask about one project, report, resource, or risk.	Specific questions produce clearer answers.
Grounding	Project graph, reports, docs, and evidence.	Cloudeval should say when evidence is missing.
Output	Answer plus next step.	Use reports for decisions that need source support.
Sharing	Export or share reviewed findings.	Do not expose private resource details unintentionally.

Who this is for

Cloud architects
SREs
DevOps engineers
Security reviewers
Engineering managers

How it works

1
Create or open a Cloudeval Azure project.
2
Make sure the project has diagram, file, sync, or report context available.
3
Ask a specific architecture question such as what changed, what is risky, or what depends on a resource.
4
Review citations, evidence, or graph insight cards where available.
5
Run missing reports when Cloudeval says evidence is incomplete.
6
Export or share findings when the answer needs stakeholder review.

Architecture Q&A over a live project

Instead of asking a generic model whether an architecture is secure, ask Cloudeval which project resources expose public entrypoints.

Then ask which report shows the finding and which details are stale or missing.
The answer should reflect the project state rather than a generic best-practice checklist.

Terminal and agent workflow

Use this command for one-shot project questions.
shell
```
cloudeval ask "Summarize this project's top architecture risks" --project <project-id>
```
MCP-compatible agents can query Cloudeval context from supported clients.
CLI sessions can preserve useful local history for automation and investigation workflows.

How to trust the result

Ask which resource, report, or diagram element supports the answer when it affects a decision.
Check sync time and report confidence before sharing recommendations.
Use narrow questions for specific resources, findings, or dependencies.

FAQ

What does Cloudeval use to answer architecture questions?

Answers can use project metadata, connection metadata, template content, diagram/resource graph data, reports, report history, pricing information, docs, and chat history where available.

Will Cloudeval answer if evidence is missing?

It should identify missing evidence and suggest a concrete next step, such as running a report or refreshing project context.

Can I ask from the terminal?

Yes.

The Cloudeval CLI includes ask and session workflows for terminal and automation use.

What is an AI cloud architecture assistant?How does an MCP server help with cloud architecture review?What can I do with the Cloudeval CLI for cloud architecture?

Build this workflow in Cloudeval

Start with a Cloudeval Azure project, then connect diagrams, reports, source links, CLI automation, and MCP-compatible agent workflows around the same source of truth.