Workflow: record your screen while narrating → get structured markdown with inline screenshots mapped to what you said.
Project principles:
• open source (MIT) • free forever core workflow • BYOK/local options
I’m not trying to build lock-in here — mostly trying to make a tool people actually use and improve. Would love technical feedback on reliability of timestamp/frame matching and report format.
Happy to answer architecture questions or share implementation details.