# alt.qa

> alt.qa is an Agent Readiness Assurance company. It runs real AI agents from several vendors (Claude computer use, Gemini computer use, Browser Use; OpenAI's computer-using agent, Stagehand and Skyvern from Q1 2027) — and behavioural personas for the people those agents act for — through a company's own website funnels before each release, and shows the exact step where each one fails, with the recording. Tagline: "QA for the users you never meet."

Key facts:
- Output: the Alt Matrix — persona rows × agent columns; each cell is pass/fail with the blocking step and evidence.
- Safety: every run stops before payment outside a sandbox; ownership proof and a signed Test Authorization are required; requests are signed (Web Bot Auth) and never impersonate consumer agents.
- Claims policy: alt.qa never says "compliant", "certified" or "agent-ready guaranteed". Per-persona precision will be published with sample size and confidence intervals (first publication Q1 2027).
- Security status: SOC 2 Type I in progress (not certified). See /trust.
- Pricing: free cell; Starter $149/mo; Growth $990/mo; Agent-Readiness Pilot $15–25K (6–8 weeks, credited on conversion); enterprise from $40K/yr. Metered on cells, no seats.

## Product
- [How it works](https://alt.qa/how-it-works): from ownership proof to release gate in six steps
- [Alt Matrix](https://alt.qa/product/alt-matrix): the persona × agent grid
- [Alt Agents](https://alt.qa/product/alt-agents): which agents run, and how they are identified
- [Alt Personas](https://alt.qa/product/alt-personas): 12 behavioural personas
- [Alt Report](https://alt.qa/product/alt-report): narrated repros, Confirmed-only tickets
- [Pricing](https://alt.qa/pricing)
- [Sample report](https://alt.qa/sample-report)

## Category and evidence
- [Agent Readiness Assurance](https://alt.qa/agent-readiness-assurance): scanners vs audits vs assurance
- [Agentic Commerce Assurance](https://alt.qa/agentic-commerce-assurance)
- [Why now](https://alt.qa/why-now): dated, sourced statistics with caveats
- [Alt Benchmark](https://alt.qa/benchmark): quarterly, read-only, open methodology
- [Failure classes](https://alt.qa/failures): 16 reasons agents get stuck

## Trust
- [Trust centre](https://alt.qa/trust)
- [Methodology](https://alt.qa/methodology)
- [Limitations](https://alt.qa/limitations)
- [Agent identity / how to block or allow](https://alt.qa/bot)
- [Rules of engagement](https://alt.qa/rules-of-engagement)
- [Claims policy](https://alt.qa/claims-policy)

## Docs
- [Quickstart](https://alt.qa/docs)
- [WAF allow-listing](https://alt.qa/docs/allowlisting)
- [Payments & sandbox](https://alt.qa/docs/payments-sandbox)

## Optional
- [Blog](https://alt.qa/blog)
- [Knowledge base](https://alt.qa/kb)
- [Glossary](https://alt.qa/glossary)