# SnT AutoTool — Product Overview

> SnT AutoTool is enterprise AI TestOps for network and security QA: it turns product intelligence and real lab topologies into executable tests, then uses actual execution evidence to verify that each outcome occurred for the intended reason.

SnT Networks builds SnT AutoTool for QA teams validating firewalls, routers, SD-WAN, SSE/SASE, ZTNA, and security platforms before release.

## Core capabilities

- **Product intelligence (DKC):** Structured capability graphs per vendor/product from documentation ingestion.
- **Topology-aware semantic tests:** Tests understand lab layout, device roles, traffic paths, and execution anchors.
- **Company curated feedback loop:** Teams upload example tests; future generation learns that company's writing style (title verbs, evidence language, step patterns).
- **AI Test Assurance:** Server-side validation that tests fail for the intended reason, paths are evidenced, and missing inputs block only affected tests.
- **Distributed execution agent:** Desktop agent runs ping, SSH, traceroute, and vendor CLI checks inside the customer's lab without exposing credentials to the cloud.
- **Upgrade regression:** Re-run the same portfolio after firmware upgrades with before/after comparison.

## Supported vendor domains

Fortinet FortiGate, Netgear business routers (PR60X/PR460X), Netskope One, Palo Alto, Cisco, Juniper, Cloudflare Zero Trust, and additional platforms via AI discovery from vendor documentation.

## Proof points from enterprise trials

- **Network appliance vendor:** 41 curated and generated tests aligned to the company's **Verify** title style and evidence-capture language.
- **Enterprise integrator:** Curated examples taught **Validate** phrasing and audit-evidence requirements for firewall, connectivity, and compliance scenarios.
- **SSE / CASB platform:** Curated examples produced **Confirm**-style titles and policy-block validation patterns.
- **AI Test Assurance:** A negative test receives `verified_expected_failure` only when the observed failure reason matches the expected reason; a wrong reason receives `unexpected_failure`.
- **Tenant isolation and input readiness:** Company-curated style remains tenant-isolated, and missing required inputs block only the affected tests.

## Typical workflow

1. Admin provisions company user accounts.
2. QA discovers/onboards product intelligence for target devices.
3. Team builds or imports lab topology (Environment Builder).
4. Semantic tests generated from topology + product knowledge + company curated style.
5. Execution agent connected on lab network; tests run with assurance verdicts.
6. Results and CSV/JSON exports include audit fields for release sign-off.

## FAQ

**What is SnT AutoTool?**
SnT AutoTool is an enterprise AI TestOps platform for network and security QA. It combines structured product intelligence (DKC), real lab topologies, topology-aware test generation, private-lab execution, and audit-ready evidence.

**How is SnT AutoTool different from generic test automation?**
Unlike generic script-only tools, SnT AutoTool combines vendor-specific product intelligence with the customer's actual lab topology and company-curated test style. AI Test Assurance verifies that failures occur for the intended reason instead of treating every failure as a pass.

**Which network and security vendors does SnT AutoTool support?**
SnT AutoTool supports Fortinet, Netgear, Netskope, Palo Alto, Cisco, Juniper, Cloudflare Zero Trust, and additional platforms. New products can be onboarded through AI discovery from vendor documentation.

**Can SnT AutoTool run tests in a private lab?**
Yes. A lightweight distributed execution agent runs connectivity, SSH, traceroute, and vendor CLI checks inside or near the customer's lab. The platform orchestrates tests and centrally captures results and evidence.

## Contact

- Product: https://auto-tool.sntnetworks.net/
- Company: https://sntnetworks.net/