Meeting minutes into NotebookLM
Claude in Chrome, on the Haiku model, extracted exact text of three years of W3C task force meeting minutes into markdown for NotebookLM.
Three years of minutes (2022-2024) were extracted in about 3 to 4 hours across three sessions and loaded as one source per year.
Overview
How it was set up
Claude in Chrome extension with the Haiku model, pointed at the minutes archive page. Output was requested as markdown in a code block, then copied into text files.
Worth knowing
It stalled after 3 or 4 months, so work was split into monthly chunks. A usage limit also hit once. Author spot-checked output manually.
A prompt to start from
Go to each page of minutes for the meetings in [years]. From each page, extract the exact text of the minutes. Do not summarize. Present the text as a markdown file in a code block.Written by Agentic Feats from the source. Not the author's exact prompt.
Sources 1
History
- Added to Agentic Feats
- 2026-10-13
- Pick
- No
- Markdown copy
- /c/meeting-minutes-into-notebooklm-claude-in-chrome.md
More in research
36 featsResultIt created the text file as asked, but after clicking a result it took a screenshot too early and hallucinated the answer.
ResultOperator handled tour search and lesson plan but was slow; adding three grocery items took 15 minutes and an Uber price check took five.
ResultThe draft was usable but weaker than accepted NSF summaries; thinking mode helped, though it began hallucinating and misreading the sources.