Why This Search Exists
Teams that automate authenticated websites often discover that hosted APIs are awkward for workflows tied to a specific person, machine, or browser state. A remote endpoint cannot safely inherit the user's current tabs, stored cookies, or transient page state without turning into a much more complex system.
That is why searches for a browser API often split into two needs: a local control layer for stateful workflows, and a hosted API for public, read-only retrieval. Mixing the two usually creates confusion.
Recommended Approach
A local browser API keeps state where it already exists. The daemon can expose browser commands on `127.0.0.1:19824`, while the browser itself remains under the user's control. That is a better fit for shell scripts, local services, and team-specific tooling running on the same machine.
iatlas-browser uses that pattern directly. You can call `/status` and `/command` over localhost, while still reserving `miaoda.vip` for the hosted subset that is safe to expose remotely.
Key Takeaways
- Use localhost for workflows bound to a person's active browser session.
- Use hosted APIs for public retrieval, not personal browser state.
- Local HTTP control is easier to integrate with scripts and orchestrators than raw CDP alone.
- A local daemon is a cleaner architecture boundary than turning every workflow into a remote browser service.
Fast Start
- Start the daemon locally with `iatlas-browser daemon`.
- Check health via `curl -s http://127.0.0.1:19824/status`.
- Send browser commands through `POST /command` using JSON payloads.
- Keep hosted `miaoda.vip` usage for public URL open requests and hosted adapters only.
Next Action
Download API examples
Move from research to implementation by choosing the correct boundary: local runtime for real-session work, hosted API for public-safe retrieval.