Portable Chat Memory Across Models (L3)
Open-source browser extension (Chat Archive) exports Claude/ChatGPT/Gemini/Grok threads to JSON or Markdown, fully local, zero outbound inference. Switch providers mid-thread without losing context. Built by John Williams (FXOps) and demoed on the GTM AI Podcast.
The steps
- 01
Install the Chat Archive extension
Tool: github.com/fxops-ai
Chromium/Edge extension. Refresh the model's tab once after install so the extension gets DOM access. All capture happens locally, nothing leaves your machine.
- 02
Export the thread you want to migrate
Tool: Extension
Pick JSON (for structured re-use) or Markdown (human-readable). Each turn is tagged user/assistant with timestamps.
- 03
Re-upload to the destination model
Tool: Claude / Grok / ChatGPT / Gemini
Drop the JSON file into a new conversation with the target model + "Pick up where we left off." The new model has the full context within seconds.
- 04
Build a personal/team knowledge corpus
Tool: Notion / vector DB
Archive exports into one folder over time. Feed into any model as RAG to build a digital twin, patterns and threads you'd otherwise forget.
Tools in this playbook
- github.com/fxops-ai
- Extension
- Claude / Grok / ChatGPT / Gemini
- Notion / vector DB
Next playbooks
Unfamiliar terms are defined in the AI and Revenue Dictionary. Related frameworks live in the framework library.
