Switching LLMs means losing years of context. Anthropic's Import Memory official prompt extracts ~40 stored facts: about 2% of what's in your conversations. Hermit processes your full ChatGPT data export and generates one structured profile per ChatGPT project and recurring theme, with temporal awareness (ACTIVE vs PAST). Ready-to-paste files for Claude Memory, Claude Projects, Gemini Gems, or any LLM. Free analytics. One-shot pricing, not a subscription. Data deleted within 24h.
Hey PH! I'm JK - builder based in Paris, working on tools that preserve memory and context, whether it's family stories before they disappear (Tirith) or AI context across platforms (Hermit).
I switched from ChatGPT to Claude after 3 years and 1258 conversations. The hardest part wasn't finding a better model - it was trying to manually migrate my context, and ended up losing most of it. Three years of projects, preferences, working style, personal context - gone the moment you switch.
Anthropic just launched the Import Memory feature, with a prompt to paste into Chatgpt. It's a great first step, but it only transfers what ChatGPT stored in memory: about 40 flat facts. My dental whitening schedule from 2024 made the cut. Some current projects didn't.
So I built Hermit. It processes your full conversations.json export through a multi-model Anthropic pipeline and generates:
→ A global behavioral profile (who you are, how you work, how to challenge you)
→ A 6-month current snapshot (active priorities, not ancient history)
→ Up to 25 topic deep-dives, one per ChatGPT project and recurring conversation theme
→ A ready-to-paste Claude Memory Import file (~47% of the 75K capacity vs ~10% from the native prompt)
→ 25 enriched memory bullets for Memory Edits
→ Output files compatible with Claude Projects, Gemini Gems, and any LLM that accepts system prompts
One tricky problem: LLMs have no concept of time. "User works at Notion" when you left a year ago. "User is apartment hunting" when you moved in 8 months ago. That's not hallucination - it's broken context.
Hermit's pipeline calculates recent activity for each theme and applies ACTIVE/PAST labels to fix this.
The free tier gives you full analytics on your ChatGPT history: conversation count, token stats, topic clustering, ChatGPT project detection. Genuinely useful on its own.
Full profiles: €19.90 one-shot. Not a subscription.
Code PRODUCTHUNT for 25% off during launch week.
On privacy: only conversation excerpts are sent to the Anthropic API (zero-retention policy, nothing stored, nothing used for training). Everything auto-deleted within 24h. No human ever sees your data. I built Hermit to migrate my own 1k+ conversations, including health stuff and personal things I would never share publicly. I wouldn't use a tool I didn't trust with my own data.
Built with Next.js 15, Supabase, Tailwind v4, Inngest, and Stripe. The pipeline is 100% Claude-powered.
🐚 Shed your shell. Keep the soul. 🦀
Recently, Openai accepted a defense deal with US DOD, the same deal which Anthropic refused bc it didn't want their ai to train for auto-"killings". Seeing this, many openai users wanted to migrate away from it due to safety concerns. Before, there was no easy way to save the ai history, settings and info. Now Hermit just allowed the trasnfer to be easy.
Weird question but did you build Hermit with that in mind? @jkjakubowski
OpenAI: "here are 40 facts about yourself"
Hermit: "actually here are 95 patterns you didn't even know you had"
me: 😳 why does an AI know me better than my therapist
Currently switching to Claude from others, let me check it out! All the best for the launch!
Oh man I went through this exact pain last year. Had like 2 years of ChatGPT conversations and switching to Claude meant losing all that built-up context about my projects and preferences. The fact that you're generating actual Claude Memory Import files is super smart — way better than just dumping raw conversation history. How accurate is the behavioral profile it generates? Like does it actually capture working style nuances or is it more surface level stuff?
Notice that this product could be useful for cases of moving from one account to another within the same provider
You plan to make the Hermit a provider agnostic product? Also make it usable for anthropic and google?
Great timing, due to political reasons and so on I heard multiple people switching between models and this definitely helps. Also when you want to compare the results of another tool but need to provide plenty of context to be able to do so. All the best of luck!
