What is new, improved and fixed in Firemark, newest first. Every change a customer would notice, with the date it shipped.
Showing MCP changes. Show everything
Find prompts where a competitor is named and you are not, where some engines name you and others do not, and where you are named without a link. A gap export and a source-gap view come with it.
Prompts that already contain your brand name get a chip on the Prompts screen, so you can tell them from open buying questions. Counts and results stay as they were.
Sources now lists which of your own pages AI answers cite, with how many answers cited each page and how it fared against competitors, per engine. Your assistant can read the same list.
Firemark's MCP server has a start_run tool. Your assistant can start a confirmed brand's first run, or retry once after a run that failed.
The last sign-up step now offers to connect your assistant, so you can set a brand up in Claude or ChatGPT instead of filling in the form. Each assistant has its own step-by-step guide.
Start tracking your brand, or open the demo to see a full report first.