Summarize a YouTuber's latest video into a Google Doc daily
A Make scenario runs at 9 AM, finds a YouTuber's newest video, has Hugging Face summarize it, and appends the summary to a Google Doc.
Comet found major LLM/agent training papers from six months with links and analysis, then saved them to a Word doc via Outlook.
Comet found major LLM/agent training papers from six months with links and analysis, then saved them to a Word doc via Outlook.
Nitika Sharma gave Comet one prompt asking for every major LLM/agent training paper from the past six months. She wanted each paper's link, publication date, an analysis, how the approaches differ, and one unique factor. She then asked it to put the results in a Word doc. For that step she connected Comet to her Outlook account. She could watch the browser agent work in the main window.
Comet returned a large body of reading material and produced the Word doc, which the author links. The post gives no paper count and doesn't say how accurate the analyses were.
Saving the doc required linking an Outlook or Google account to Comet. Comet is free on Windows and macOS as of October 2025.
Find the most important papers on [research topic] from the last [number] months. For each, give the title, authors, link and a two-sentence summary of why it matters. Compile everything into a Word document titled [title] and tell me where it was saved.
A Make scenario runs at 9 AM, finds a YouTuber's newest video, has Hugging Face summarize it, and appends the summary to a Google Doc.
An n8n workflow scrapes Reddit posts and comments via Bright Data, extracts concerns and themes with Gemini, and writes reports to Notion.
A user left a Gemini agent running overnight to analyze 170k LLM logs with the llm CLI and a free model, and it finished the job despite limits.