SkillVaultskills Browse all 500 skills

Networking · Version 1.6.0 · Reviewed 2026-08-02

VPC Network Design Reviewer

Make a defensible decision about subnet planning and egress path cost with evidence, explicit trade-offs, and a verification plan.

4 method steps 5 documented failure modes 4 diagnostic checks 7 quality gates

Reviews subnetting, routing, egress paths, and peering for cost, reachability, and blast radius.

₹149 one-time

Get this skill archive

What this skill helps you do

  • Subnet planning
  • Egress path cost
  • Peering and isolation

How VPC Network Design Reviewer works

You provide

Failing request path, timings, and intermediary configuration

It inspects

Layer isolation for subnet planning: resolve, connect, handshake

It decides

A egress path cost fix attributed to a specific hop

You verify

Reproduce from the affected path, not from a healthy one

What it checks first

VPC Network Design Reviewer reviews subnetting, routing, egress paths, and peering for cost, reachability, and blast radius. Use it when the work involves Subnet planning, Egress path cost, Peering and isolation.

  1. Which layer the failure occurs at, since a name-resolution failure and a TLS failure look similar to the application.
  2. Whether the failure is symmetric, because one-way reachability points at policy rather than transport.
  3. Timing: connection setup, first byte, and transfer are separate problems with separate causes.
  4. Whether intermediaries such as proxies, load balancers, or service meshes terminate the connection.

Failure modes it recognizes

  • DNS search-path expansion adding latency and intermittent failures for external names.
  • An idle timeout on an intermediary silently closing connections a pool still believes are open.
  • MTU mismatch causing large payloads to hang while small requests succeed.
  • TLS SNI or certificate mismatch surfacing as a connection reset with no application log.
  • A policy denying return traffic, producing a timeout rather than a refusal.

Answers it will reject

  • Blaming the network before confirming resolution, routing, and policy each independently.
  • Increasing timeouts to mask an intermediary that is closing connections deliberately.
  • Testing from a different host or network than the one experiencing the failure.

Decision rules it applies

  • Isolate the layer before proposing a fix: resolve, connect, handshake, then request.
  • Reproduce from the affected network path; a successful test from elsewhere proves nothing.
  • Prefer connection-level evidence over application logs when the failure is pre-response.

Evidence it asks for

  • Capture resolution, connect, TLS, and first-byte timings separately.
  • Test with and without the intermediary to attribute the reset.
  • Confirm reachability in both directions rather than assuming symmetry.

The method inside

  1. Map the artifact, actors, boundaries, and invariants relevant to subnet planning.
  2. Trace concrete failure or abuse paths for egress path cost; do not report checklist items without a mechanism.
  3. Prioritize peering and isolation findings by impact, likelihood, confidence, and cost of correction.
  4. Recommend the smallest defensible change, then define how an independent reviewer can verify it.

Deliverables

  • Subnet planning assessment
  • Egress path cost decision and action plan
  • Peering and isolation verification checklist

Evidence requirements

  • Functional and quality requirements
  • Scale, latency, consistency, cost, and compliance constraints
  • Current topology and alternatives considered

Quality gates

  • Every material claim traces to supplied evidence or is labeled as a hypothesis.
  • The response follows the declared deliverable contract.
  • No execution, access, measurement, or verification is invented.
  • Secrets and personal data are redacted rather than repeated.
  • The user receives a concrete independent verification step.
  • The relevant failure modes in this domain were considered rather than only the reported symptom.
  • No listed anti-pattern was recommended as a solution.

Example task

Input

Review our VPC design. We are running out of IP space and our NAT gateway costs more than our compute.

Expected output

Both problems share a cause: everything sits in one flat network with a single egress path. Re-plan CIDR allocation with room per environment, and route service traffic through private endpoints so it never touches NAT, which is where the cost is accumulating...

Boundaries and compatibility

Ideal for

  • Subnet planning: produce a decision or artifact grounded in supplied evidence.
  • Egress path cost: produce a decision or artifact grounded in supplied evidence.
  • Peering and isolation: produce a decision or artifact grounded in supplied evidence.

Out of scope

  • Producing a generic reference architecture without requirements
  • Hiding material trade-offs behind best-practice language

Agent compatibility

  • GitHub Copilot custom agents
  • Claude Agent Skills / SKILL.md
  • Any instruction-following chat model

Tool policy: Advisory by default. No tools are assumed. If the host provides tools, use read-only evidence gathering unless the user explicitly approves a scoped write or execution action.