Notes and limits

API request limits apply per plan tier. The number of API calls you can make is governed by your plan, so design clients to handle rate limiting and back off when asked.

The browsers run on your own hardware, locally or on a server you host. Because nothing renders in a metered cloud, there is no per browser hour charge on any plan. The practical limit on concurrent sessions is your own CPU, memory, and proxies.

• stealth: "off" is Chrome only. Firefox engine profiles always spoof their fingerprint.
• automation and ephemeral are accepted but advisory today. They do not yet change launch behavior.
• Per launch overrides apply only to that launch and are never written back to the saved profile.
• Read the port and token from Settings rather than hard coding them, since they can differ per machine.