Three typical integrations, step by step.
Send sourced articles to a headless CMS, run an approved publishing pipeline, feed your dashboards. Each scenario sets out who does what, on the Aive side and on yours. The exact code lives in the public documentation.
How do I integrate Aive with my existing system?
Most integrations follow the same pattern: your system submits a video, Aive prepares analyses and articles linked to the source moments, a webhook signals that an output is ready, your team approves, then your system retrieves the export. What changes from one scenario to the next is the destination: a CMS, a publishing chain or an analytics tool. Each destination is subject to your configuration.
- Headless CMS
- A content management system that exposes content through an API, without imposing a front end.
- Enabled destination
- A target system explicitly configured and authorised to receive Aive outputs.
Pick your starting point.
Headless CMS
Prepared articles land as drafts in your CMS, with their sources and video moments. Publishing stays in your CMS.
Publishing pipeline
A webhook triggers your review chain; only an approved version goes to an enabled destination.
Analytics export
Exports feed your data warehouse or dashboards, with their source identifiers.
Aive side, your side, human side.
| Scenario | Trigger | Aive side | Your system | Human control |
|---|---|---|---|---|
| Headless CMS | Video submitted by your system | Analyses and draft article, linked to moments | Draft created in the CMS | Review and publishing in the CMS |
| Publishing pipeline | Webhook: output ready | Versions prepared with their status | Routing to the review queue | Approval of version and destination |
| Analytics export | Scheduled on your side | Structured export with source identifiers | Loaded into the warehouse or dashboard | Reading and interpreting results |
Descriptive scenarios. Exact paths, parameters and fields are in the public documentation at api.aive.tech.
Publishing pipeline, step by step.
- 01Submit the videoYour system sends the video or its reference, with language, owner and rights.
- 02Wait for the webhookNo polling: the notification tells you the outputs are ready.
- 03Retrieve the outputsAnalyses and draft articles, each linked to its source moments.
- 04Route to reviewYour workflow tool assigns the review to the right person.
- 05Publish the approved versionOnly the approved version goes to an enabled destination; the action is logged.
Why no code here?
An approximate snippet would mislead. The cURL examples in the public documentation are generated from the specification and stay in step with it.
- Official cURL examples
- OpenAPI 3.1.0 specification
What these examples do not cover.
- No connector is presented as enabled by default: each destination is validated connector by connector.
- These scenarios do not replace the API reference; they reproduce none of its paths or fields.
- An analytics export describes observed results; it establishes no causal link without a dedicated protocol.
- Availability of each scenario depends on your contract, rights and configuration.
Questions about integrations.
Where can I find ready-to-use code?
In the public documentation at api.aive.tech: the cURL examples there are maintained alongside the OpenAPI 3.1.0 specification. This page describes scenarios, not code.
Is my CMS compatible?
Compatibility is validated connector by connector: tested capabilities, version, rights and contractual availability. The CMS integrations page sets out the approach.
Can I publish without human approval?
The recommended path assumes approval before publishing. Any further automation is scoped in the contract, with logging and the ability to roll back.