Get a 1-page QA report on your app.
Tell us where to look. Our team runs an exploratory session, reviews your test inventory, and sends back a prioritized list of what to fix and what's already solid. Complete audits are requested by email to team@goqa.ai.
What you get
- Heuristic evaluation against 10 standard quality dimensions
- Top 5 bugs ranked by severity, with reproduction steps
- Coverage gap analysis — where you're under-tested
- 30-min walkthrough call with one of our QA leads
- No obligation. We don't pitch; we leave you the report.
Request your audit
No credit card required.
Complete audits are requested by email. Send the site or app URL, what you would like audited and your stack to team@goqa.ai, and we reply by email to agree the scope.
Request a complete auditThe button opens your email app with the request started. Never send passwords, API keys or customer data by email.
AI readiness assessment
For teams shipping a chatbot, RAG app or AI agent who want evidence of how it behaves before release.
AI readiness assessments are scoped and confirmed with you before testing begins. Depending on the agreed engagement, GoQA may begin with URL-visible behavior and recommend API or trace integration for deeper evidence.
What it can cover
- Answers checked for unsupported claims, citation support and appropriately stated uncertainty
- Refusal quality and resistance to defensive prompt-injection and data-disclosure probes (benign proxies only)
- Robustness to paraphrases, typos and ambiguous questions
- For agents: visible behavior, and — once traces are connected — tool use, approvals and whether completion claims match tool results
What to provide
- The URL of the chat page, and whether it needs a login (URL-based testing stops at login walls and CAPTCHAs)
- What the app is (chatbot, RAG app or agent) and what it should and should not do
- Confirmation that you own it or have written permission to test it
- Whether an OpenAI-compatible API or agent traces could be connected later
Summarize these briefly in an email to team@goqa.ai: start an AI readiness request with these prompts filled in. Never send passwords, API keys or customer data by email.
| Aspect | Website QA audit | AI readiness assessment |
|---|---|---|
| Looks at | Your app's pages and flows: bugs, quality dimensions and test coverage gaps | What your LLM app or agent says and does in response to test conversations |
| Access | Where to look in your app | Agreed with you during scoping; may begin with URL-visible behavior, with API or trace integration recommended for deeper evidence |
| Outcome | A 1-page QA report (see “What you get” above) | Risk findings and test evidence within the agreed scope — not a certification |
URL-only limitation: URL-only testing observes externally visible behavior. It cannot prove hidden prompts, temperature, retrieval documents, tool calls, guardrails, memory, or internal reasoning. Findings describe observable behavior and potential risks. Connect an API or trace source for deeper evidence.
The AI Assurance workflow
The product workflow an assessment draws on. Which steps apply is agreed with you during scoping.
Step 1 of 6: Authorize the target
Step 2 of 6: Select an approved plan
Step 3 of 6: Run bounded tests
Step 4 of 6: Review evidence and findings
Step 5 of 6: Remediate and retest
Step 6 of 6: Apply a release gate
In order: authorize the target → select an approved plan → run bounded tests → review evidence and findings → remediate and retest → apply a release gate.
Three levels of evidence
Each level adds evidence and the workflow stays the same. URL-only testing is not available yet; connect an API or send traces.
Evidence depth 1 of 3
URL-only (black-box)
URL-only testing is not available yet. When released, GoQA will interact with your chat page only through the visible interface, within the scope you declare.
Can show
- Visible answers to bounded, defensive test prompts
- Refusal quality and disclosure behavior shown in the UI
- Visible citations, error states and response timing
Evidence depth 2 of 3
API-connected
GoQA calls your OpenAI-compatible chat endpoint directly. Credentials are encrypted and never shown again.
Can show
- Exact model responses with token usage, cost and latency
- Configuration sensitivity (temperature, top_p, max tokens, seed)
- Structured output and schema checks
Evidence depth 3 of 3
SDK / trace-connected
Your server sends run, retrieval, tool, agent-step, latency, token, cost and error events through the GoQA SDK.
Can show
- Full agent trajectories: tool selection, arguments, order, approvals and delegation
- Completion claims checked against recorded tool results
- Loops, budgets, errors and recovery from reported events
This is a readiness assessment with risk findings and test evidence. It is not a compliance certification and does not guarantee that a system is safe. Policy packs are technical test packs, not legal advice. Read the AI Assurance docs.