The whole loop is here — write the prompts, run them across models, read the transcripts, see which competitors and which sources came up — so a claim about brand visibility can be checked instead of argued about. The history call is what turns a snapshot into a trend.
A management and reporting surface for LLM Brand Monitor: brand monitoring projects and the prompts inside them, scans run across models, and the results, transcripts, competitors and citations those scans produce.
- Projects listed, read with their prompts and models, created, updated and archived
- Prompts added to a project or removed from it
- Scans started across models, with scan status readable while one is in flight
- Results listed per scan, and the raw transcript of a model's answer
- Competitors named in the answers, and the links the models cited
- History over time, so a change in how you are mentioned can be seen rather than inferred
- The available model list — 350+ of them — and your own usage against quota
An LLM Brand Monitor account and an API key from your profile page, beginning lbm_ and passed as LBM_API_KEY. Run it with npx @serpstat/llm-brand-monitor-mcp; no local build needed.
One command plus a key — npx -y @serpstat/llm-brand-monitor-mcp, then supply credentials