# VeroniQA

> VeroniQA is an AI assistant for API testing, built on AgentQA. Give her your requirements, documents and links; she answers with cited sources, generates and runs API tests, and explains what she found.

VeroniQA is a proof of concept by Rowan Kendal (https://www.rowan-kendal.com/). The public demo at https://veroniqa.vercel.app/ uses
simulated models and runs tests only against a bundled sample Orders API with planted bugs. No
account, sign-up or API key is needed: open https://veroniqa.vercel.app/talk/ and a private workspace is created.

## Links

- [Home page with the introduction video](https://veroniqa.vercel.app/index.md): Markdown; also served at https://veroniqa.vercel.app/ to
  clients that send `Accept: text/markdown`
- [Talk to VeroniQA (the app)](https://veroniqa.vercel.app/talk/): an interactive web app (it needs a browser with
  JavaScript); request it with `Accept: text/markdown` for a description
- [Developers: API guide, quickstart, errors, rate limits, CLI](https://veroniqa.vercel.app/developers.md)
- [OpenAPI 3.1 description of the public API](https://veroniqa.vercel.app/openapi.json)
- [API catalog (RFC 9727)](https://veroniqa.vercel.app/.well-known/api-catalog)
- [Privacy](https://veroniqa.vercel.app/privacy.md)
- [Security contact](https://veroniqa.vercel.app/.well-known/security.txt)
- [Source code: AgentQA and VeroniQA](https://github.com/Zapkid/agentqa)
- [README](https://github.com/Zapkid/agentqa/blob/main/README.md)
- [Security notes](https://github.com/Zapkid/agentqa/blob/main/docs/SECURITY.md)
- [Known limitations](https://github.com/Zapkid/agentqa/blob/main/docs/LIMITATIONS.md)

## Features

- Answers with sources: Ask about your API and get answers grounded in your own documents, with numbered citations. Citations that do not match a retrieved passage are dropped, and "not found" is a real answer.
- Tests from your requirements: Say “run the tests”. AgentQA plans risk-ranked tests from the spec and your documents, runs them, and triages each failure with evidence.
- Knowledge you can trust: Upload Markdown, text, HTML or PDF, or add links. Links are fetched from public addresses only, and text that looks like a prompt injection is quarantined.
- Retrieval that fits the domain: Each project has its own vector store (Chroma) combined with keyword search (BM25), so field names and error codes are found as reliably as plain prose.
- Cost-aware by design: A cheap model routes each message; stronger models are used only where they change the result. Every step is traced with its tokens and cost.
- Fits how teams work: Use the browser, let coding agents call AgentQA over MCP, or gate pull requests on an eval and a performance smoke test in CI.

## API and CLI

A free, keyless sandbox API at https://veroniqa.vercel.app/api (OpenAPI: https://veroniqa.vercel.app/openapi.json). `POST /api/v1/ask` with
`{"question": "..."}` returns a cited answer from the demo API's requirement documents;
`GET /api/v1/results` returns the measured results. Errors are RFC 9457 problem documents
(`application/problem+json`) with a `code` and a `hint`. The official CLI is `veroniqa`
(`pipx install "git+https://github.com/Zapkid/agentqa#subdirectory=clients/veroniqa-cli"`).

## Frequently asked questions

- What is VeroniQA? VeroniQA is a chat assistant for API testing. She keeps a project's requirements in a searchable knowledge base, answers questions with citations, and starts AgentQA test runs that plan, generate, execute and triage API tests.
- What is AgentQA? AgentQA is the engine behind VeroniQA: an observable, guarded multi-agent system that turns an OpenAPI spec and requirement documents into risk-ranked tests, a run, triaged findings and an executive summary.
- Does the online demo use a real AI model? No. The public demo uses deterministic simulated models, so it needs no API keys and costs nothing to try. Answers are extracts from the documents. Run it yourself with your own keys for real model answers.
- Is what I upload private? Each visitor gets a private, temporary workspace tied to their link. Workspaces are deleted after 12 hours or when the server restarts. Do not upload confidential documents to the demo.
- Can I test my own API? In the demo, tests run only against the bundled Orders API, which has planted bugs to find. To test your own API, run VeroniQA from the GitHub repository.
- Is there an API or a CLI? Yes. The public API is a free, keyless sandbox described by an OpenAPI 3.1 document at /openapi.json, with a guide at /developers. The official veroniqa command-line client calls it.
- Is it production-ready? No. It is a proof of concept that shows the approach end to end. Fitting it to a specific team's APIs, models and controls comes next.

## Video transcript

Meet VeroniQA, your AI assistant for API testing. She turns your requirements into tests that run, and results you can trust. Start a project, and give her what your team already has: requirement documents, PDF files, or links to your docs. Every source is indexed, and scanned for prompt injection before she ever reads it. Ask a question in plain English. VeroniQA answers from your own documents, and shows the passages she used. If the answer isn't there, she tells you, instead of guessing. Links are fetched safely. Internal and cloud-metadata addresses are refused, so your knowledge base can't be turned against your network. When you're ready, just say: run the tests. VeroniQA plans, writes and runs API tests against your service, then triages every failure, with evidence. In this demo, she found both planted bugs, with 46 tests passing. Behind her, the AgentQA engine gives easy work to cheaper models, and checks every result. On our simulated benchmark, it found 94% of the planted bugs, at 47% of the cost. VeroniQA. Ask, add knowledge, and run the tests.