minimi 2.0

98 comments were scraped, but quality is split. The relevant signal comes from ~14 ProductHunt launch comments; most Lemmy comments discuss EU chat-control legislation, coffee/health topics, and a Wayland window-focus bug — none mention Minimi, so they read as keyword-collision noise. Among relevant comments, sentiment is strongly positive but skewed by launch-day hype and builder participation (one commenter states 'It's all local-first. Your context gets embedded and stored on your Mac, retrieval runs locally, and Claude pulls it over a single MCP connector. Nothing leaves your device. I know because I built it'). Reported strengths: it eliminates the 're-explain yourself every session' problem with Claude by passively 'reading what you read, hearing what you hear'; one user says it 'remembers weeks of conversations' and answers 'What are the tasks I should handle urgently' and 'Who all did I talk to today' with names, platform, and context; another cites recall of offhand details ('What should I get Jay for his birthday... it remembers the offhand thing he wanted three weeks ago'). The 'you-pick-what-it-sees' granular per-app control is explicitly praised. The sharpest pushback is technical rather than hostile: (1) memory quality over time — 'the harder test is what happens after weeks of accumulated context... what's still true versus what's become outdated'; (2) signal vs noise — risk of 'feeding Claude confidently-irrelevant context' from passive capture; (3) the deletion path — 'when I revoke an app or delete a memory, do the vectors already sitting in the local store actually go? That's the real privacy question'; (4) macOS-only, with 'any Windows roadmap?' called a main blocker. One user frames it as 'a crazy upgrade' from the same team's earlier product Shram. No independent negative reports exist in the data, but no long-term usage evidence exists either — everything relevant is from launch day.

No aggregate ratings were found for this product during the last harvest.