chat-connectors

Sync ChatGPT and Claude conversation history into a gbrain knowledge base incrementally.

29.4k|4.4k|Updated Apr 5, 2026
One-click install
npx skills add https://github.com/garrytan/gbrain --skill chat-connectors
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: chat-connectors
Source: https://github.com/garrytan/gbrain/tree/main/plugin/skills/chat-connectors
Command: npx skills add https://github.com/garrytan/gbrain --skill chat-connectors

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Manually exporting and importing AI chat history is repetitive and quickly goes stale. This Skill connects your ChatGPT or Claude account directly and keeps your conversation archive continuously synced into your gbrain knowledge base, so every conversation becomes searchable, cross-linked memory without repeated manual exports.

Core Features & Use Cases

  • Live Account Connection: Authenticate ChatGPT or Claude via a browser session cookie (stored locally at 0600 permissions) or best-effort OAuth, with a probe verifying credentials before saving.
  • Incremental Sync: A durable per-provider watermark plus a 7-day trailing window fetches only new or edited conversations; content-hash idempotency makes re-runs free and safe.
  • Scheduled Automation: Opt-in daily auto-sync per provider via autopilot (launchd/systemd/cron) or a host cron line, gated on valid credentials.
  • Use Case: You want every ChatGPT conversation automatically filed under conversations/chatgpt/ each night. Connect your account once, enable connectors.chatgpt.auto_sync, and the brain stays current with redaction, slugging, and dedup handled by the native ingest pipeline.

Quick Start

Connect my ChatGPT account and run a full sync of my conversation history into the brain.

Frequently Asked Questions about chat-connectors

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I sync my ChatGPT history into gbrain automatically?

Authenticate with `gbrain connectors auth chatgpt --cookie -` and paste your browser Cookie header, then run `gbrain connectors sync chatgpt --full`. Enable `connectors.chatgpt.auto_sync` and install autopilot for daily scheduled syncs.

How do I connect my Claude account to import conversations?

Copy the sessionKey cookie value from your browser's DevTools (Application → Cookies), then run `gbrain connectors auth claude --cookie -` and paste `sessionKey=<value>`. A probe verifies the credential before anything is saved.

What is the difference between chat-connectors and conversation-archive?

chat-connectors is the live account-connected lane that pulls conversations directly from ChatGPT or Claude on an ongoing basis. conversation-archive handles one-time imports of downloaded export files and all retrieval queries like tracing when you first discussed a topic.

Does chat-connectors support Perplexity or other providers?

No. Only ChatGPT and Claude have live connectors. Perplexity users must use the conversation-archive skill's manual conversion path with an exported file instead.

Why does my ChatGPT sync fail with a forbidden error?

A `forbidden` result means a Cloudflare or bot challenge is blocking server-side fetches from your machine. Do not retry in a loop; fall back to the official export via Settings → Export data and ingest the file through conversation-archive.

Where are my ChatGPT cookies stored and are they secure?

Credentials live only at `~/.gbrain/connectors/<provider>.json` with 0600 permissions inside a 0700 directory, never in the database, config, or any MCP payload. `gbrain connectors status` shows provenance and expiry without revealing the value.