Skip to main content

chatgpt-to-claude-mcp

Turn your ChatGPT export into a personal context profile for Claude and GitHub Copilot.

Instead of manually writing a system prompt, this tool mines your actual conversation history — your word patterns, recurring topics, and habits of thought — and distils them into a ~3k character profile that meaningfully changes how an AI responds to you.

How it works

The pipeline runs entirely locally. It never sends your conversation data anywhere.

  1. Stream conversations-*.json using ijson — never loads the full file into memory
  2. Extract user-only messages (assistant responses are just mirrors of your input)
  3. Score lines by a weighted blend of word frequency relevance (60%) and Shannon entropy (40%), with context window expansion (±0–3 surrounding lines)
  4. Surface the top proper nouns from your history — names, projects, companies — and ask which to exclude as noise (old employers, irrelevant contacts, etc.)
  5. Re-run with updated banned phrases for a cleaner signal
  6. Synthesise a profile via the LLM from the dense signal text
  7. Install globally at ~/.config/profile.md with symlinks for Claude (~/.claude/CLAUDE.md) and Copilot (~/.copilot/copilot-instructions.md)

Banned phrases are persisted to {export_dir}/banned_phrases.json so they survive reruns.

Requirements

  • Python 3.11+
  • uv (recommended) or pip

Add to Claude Desktop

~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "chatgpt-to-claude": {
      "command": "uvx",
      "args": ["chatgpt-to-claude-mcp"]
    }
  }
}

Then restart Claude Desktop and say:

"Build my profile from my ChatGPT export at ~/Downloads/chatgpt-export"

Add to GitHub Copilot (VS Code)

.vscode/mcp.json:

{
  "servers": {
    "chatgpt-to-claude": {
      "type": "stdio",
      "command": "uvx",
      "args": ["chatgpt-to-claude-mcp"]
    }
  }
}

Install via pip

pip install chatgpt-to-claude-mcp
chatgpt-to-claude-mcp

Tools

process_export(export_dir, banned_phrases?)

Runs the full pipeline. Returns:

  • top_proper_nouns — review these with the user, flag irrelevant ones
  • dense_signal — scored excerpts for profile synthesis
  • synthesis_prompt — prompt to use when synthesising the profile

save_profile(profile_content, export_dir)

Writes the synthesised profile and installs symlinks.

Getting your ChatGPT export

Settings → Data Controls → Export Data. You'll get a zip with conversations-*.json files.

License

MIT

Metadata

Release files for chatgpt-to-claude-mcp 0.1.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for chatgpt-to-claude-mcp 0.1.0
File Size Uploaded Provenance Actions
chatgpt_to_claude_mcp-0.1.0.tar.gz 11.9 kB With provenance Details for chatgpt_to_claude_mcp-0.1.0.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for chatgpt-to-claude-mcp 0.1.0
File Interpreter ABI Platform Actions
chatgpt_to_claude_mcp-0.1.0-py3-none-any.whl Python 3 none any Details for chatgpt_to_claude_mcp-0.1.0-py3-none-any.whl

Total release size: 25.0 kB

Release files / chatgpt_to_claude_mcp-0.1.0.tar.gz

Download URL chatgpt_to_claude_mcp-0.1.0.tar.gz
Size 11.9 kB
Tags Source
SHA-256 checksum
How to use checksums
3cbd2b3a06c5f458f67351a82cbb3669312d2eaa73e758d21952597758b18a43
BLAKE2b-256 checksum
How to use checksums
220a50ca6873f418280d6728dc5ed9e9751ad1d665c57d636977c2c1223e2d0c
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Mar 4, 2026.

Transparency log

Release files / chatgpt_to_claude_mcp-0.1.0-py3-none-any.whl

Download URL chatgpt_to_claude_mcp-0.1.0-py3-none-any.whl
Size 13.0 kB
Tags Python 3
SHA-256 checksum
How to use checksums
55a4f203c96e7357d0777ab946f5d57d81a912215b0cb7c91eb5376f9c316cde
BLAKE2b-256 checksum
How to use checksums
1bfb532f8727886d42b8da41c6912fea6fe1f0c710dd00fc900a92ba70a0ef41
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.7

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Mar 4, 2026.

Transparency log

Security for chatgpt-to-claude-mcp 0.1.0

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

All 2 files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for chatgpt_to_claude_mcp-0.1.0.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Mar 4, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.1.0 This release

2 release files Full provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page