The Lastfind agent answers questions about your project with the same numbers you see in the app, turns them into charts and reports, and writes SEO content checked against the pages that rank today. It learns your brand through the Brain and improves with your feedback.
Ask “which competitor grew most in ChatGPT this month?” and get the answer from your reports: visibility, brands, topics, sentiment, Semantic Branding, products, sources and AI searches. Attach a PDF, a spreadsheet or an image for context.
Ask for a chart and it lands in a report you can edit: bars, line, donut, table or key numbers from any report, with its filters. The numbers come from Lastfind, not from the AI, and can update by themselves each time the report opens.
The agent reads the top 3 Google results for your keyword, plans the outline around what they cover and miss, and writes. The Overview tab compares your text with each of them, and where it falls behind, the agent revises it.
Title tag with a 60-character count, meta description up to 155, slug, schema, internal links, images and pending items sit in their own block, outside the article. The agent fills them; you edit and they save.
About, audience, products, tone of voice, facts and what to avoid. A wizard reads your site and drafts each file with the facts it found for you to review. Memory keeps your preferences, decisions and lessons across conversations.
Fourteen ready skills for SEO and AI visibility, plus your team’s own. A thumbs down with a reason, or a correction in the chat, improves them. Every change shows under the answer with Undo, and every version stays in History.
Every content piece has an Overview tab with a table of your text against the 1st, 2nd and 3rd Google results for the keyword. Lastfind counts the words on each page; the agent judges which topics and Google questions each one covers.
Where your text falls behind the best of the three, the agent revises it, up to two rounds, and notes in the table what is still behind and why. When you edit the text, the counts follow and the table tells you the coverage needs a new check.
The agent, with its Contents, reports and Brain, is part of the paid Cloud plans and of the independent installation. The Free account does not include it.
The numbers in its answers and charts come from your Lastfind reports. When the agent saves a chart that cannot be drawn from your data, Lastfind sends it back with the reason and saves nothing.
Changes to prompts, topics and other project data wait for your approval on screen, and a viewer can only read. Changes to skills, the Brain and Memory apply right away, show up in the chat and can be undone.
No. The table shows how your text covers what the current top 3 cover. Rankings depend on much more than the text, so treat it as a quality check, not a promise.
A thumbs down with a reason, a thumbs up or a correction in the chat can update a skill, the Brain or Memory. The agent only changes a skill used in that conversation, quoting your words, and an automatic change that gets poor ratings in its first uses is undone by itself.
Yes. Connect Claude, ChatGPT, Cursor or another MCP client in Settings › AI apps. It reads the same reports as the app and makes only the changes you allow.
