Identify the exact connection attempt
Record the account, Region, instance, source location, destination and protocol. Confirm whether the intended path is public, private or through an approved access service. A timeout and an authentication rejection should not automatically be treated as the same fault.
Define which AWS resource observations your configured tools can retrieve and which host checks are available. The developer guide confirms AWS tooling, but does not establish turnkey coverage of every EC2 or VPC diagnostic operation. Verify the specific API operations before enabling the workflow.
Review cloud controls and host observations together
AWS documentation separates connectivity and authentication causes. Use that distinction to collect relevant evidence without broadly opening network access.
| Area | Review question |
|---|---|
| Identity and health | Is this the intended account, Region and running instance? |
| Address and path | Does the selected address match the intended access path? |
| Cloud network controls | Do available route, security-group and network-ACL observations explain the flow? |
| Guest operating system | Is the required service reachable and listening according to approved host checks? |
| Authentication | Does the failure occur after reaching the service, and are the access details correct? |
Illustrative example: a private instance cannot be reached directly
An engineer attempts to connect to an instance using a path that does not match the documented private-access design. The workflow starts by checking resource identity and the intended connection method, then gathers the available cloud and host evidence.
The result explains the mismatch and missing checks for review. It does not propose public exposure as a default repair. This is an illustrative scenario rather than a customer implementation or a claim that NetPrompt discovers every cloud dependency automatically.
Validate permissions and unavailable-evidence behavior
Use scoped credentials for the required observations and check the configured account and Region. A failed permission check is a coverage gap, not proof that a resource or route is absent. Preserve tool errors in the investigation record.
Where a required operation is unavailable, extend the workflow only through a supported, tested integration or an explicit engineer-provided evidence step. This page does not claim a native AWS Reachability Analyzer integration or automatic VPC topology simulation.
Require a service test after an approved action
Review the target and scope before any change to security rules, routes, instance state or host configuration. Use the applicable approval and change process, then repeat the original connection and service test.
Evaluate a known network-access fault, an authentication failure and an access-denied API response. The workflow should distinguish these cases and state the evidence behind its conclusion. Measure operator corrections and investigation effort instead of assuming that every successful API response means the service recovered.
Common questions
- Can NetPrompt automatically fix any EC2 connection failure?
- No universal coverage is claimed. Validate the required AWS operations, permissions, host access and diagnostic workflow for the selected environment.
- Does hosting NetPrompt in AWS automatically grant access to resources?
- No. Execution access requires the appropriate configured credentials, permissions and connectivity. Deployment location and operational access are separate choices.
- Should I open SSH or application ports to everyone to test access?
- Use an approved, narrowly scoped access path. This workflow does not require broad public exposure as a troubleshooting step.
Continue exploring
Further technical reading
External references explain general practices; they are not endorsements of NetPrompt AI.