Advanced

Integrations

How to connect ArtificialQA with your AI agents, import/export data, and consume results from your pipelines.

Integration types

Today ArtificialQA integrates with your ecosystem in three directions:

🤖

AI agent connection

Via HTTP/API or browser (Playwright). It's the "what gets tested".

📤

Import / Export data

Import test cases from Excel/JSON. Export reports to PDF.

🔌

REST API

Trigger runs and read results from your own scripts or pipelines.

1. AI agent connection

HTTP / API (Pro and Enterprise)

For AI agents exposing their own endpoint. The cleanest and most efficient way when the AI agent has an API.

Supports Bearer Token, API Key (in header or query), Custom Headers.
Body templating: you indicate where the user message gets injected.
Response mapping: you indicate which field to extract the bot response from.
Works with any framework (LangChain, LlamaIndex, Rasa, etc.) as long as they respond over HTTP.

Browser / Playwright (all plans)

For AI agents embedded in webpages. ArtificialQA runs headless Chromium on its cloud workers:

Supports embedded chat widgets (including commercial platforms).
Supports pages with login: define the Login Steps sequence.
Runs headless on our cloud workers.
Stores in the Run snapshot the captured HTML, selectors used, and DOM response, for later inspection.

2. Import / Export data

Import test cases

Excel (.xlsx): download the template, fill it in, upload it. Ideal when the product team hands you the cases.
JSON: useful for automation — a script generates the JSON and uploads it via UI or API.

Imported cases land in the review view before being incorporated, just like AI-generated ones: you can edit them and decide whether they go to your Test Cases catalog, to a specific Test Suite, or get discarded.

Export reports

PDF — from the Evaluation Report, PDF Report button.
API — consume structured results in JSON.

3. REST API

Available on the Enterprise plan. The REST API lets you operate the platform from your own scripts or pipelines.

Typical use cases

Trigger a run every time the AI agent is deployed.
Read the run result and block the deploy if the score falls below a threshold.
Consolidate runs from multiple AI agents in an external dashboard.
Programmatically generate test cases and upload them.

Main operations

List and create Test Cases.
List and create Test Suites.
List Test Plans, trigger a Run, query its status.
Get results and scores from an evaluated Run.

📚 Detailed API documentation. The technical reference with all endpoints, schemas, and examples is being consolidated for publication. In the meantime, reach out via artificialqa.com and we'll get you access.

What's on the roadmap

To be transparent: there are integrations not yet available that we receive as frequent requests:

Webhooks to notify Run completion.
Native Slack / Teams integration.
Native Jira or other ticket-tracker integration.
Official SDKs (Python, Node.js).

If any of these is a blocker for you, reach out — the roadmap is prioritized with real customer feedback.

Next step

If you'll be handling sensitive data or have compliance requirements, the Security & compliance section is the next stop.