Part leadership across all sessions. Gold lines mark sessions (click to open); blue lines mark memories (click to scroll).
Paste a data.json here and hit Apply. Use this to restore a backup or import an update from your AI after a journalling session.
Downloads the same encrypted file that GitHub sync uses. Safe to store anywhere: AES-256-GCM encrypted with your password and unreadable without it. Use as an offline backup or to move to a new device.
⚠ These exports contain your private data in plain text. Store them securely and do not share them casually.
Compact snapshot for sharing with your AI. Includes all parts, system info, sessions as summaries, and self moments. Chat message logs are stripped. Use the full export for device migration.
Exports all app data including full chat message logs. Use this to move between devices or browsers. Do not paste this into an AI chat.
Your mind is made of parts: the one that pushes forward, the one that shuts down, the one that numbs out when things get hard. team-sheet helps you name and understand them. No prior experience needed.
team-sheet is free, open source, and always will be. If it has been useful to you, a small contribution helps keep it going.
team-sheet is a supplement to IFS therapy, not a replacement for professional care. If you are struggling, please reach out to a qualified therapist or a service like Befrienders Worldwide.
Your AI infers which parts were present, how relationships shifted, and whether any Self-led moments occurred. Lock any field to protect your own writing from being overwritten.
Before each new session, Export your latest data.json and paste it into your AI project so it has your current state.
The Chat button walks you through setup the first time you open it.
When set, all your data is strongly encrypted in the browser. Use a long random passphrase, or generate one with a password manager.
⚠ If you forget your password, your data cannot be recovered. Keep an export of your data.json as a backup.
This deletes everything: your parts, sessions, balance history, and settings. An undo option will appear for 60 seconds.
Back up your encrypted data to a private GitHub repo. Pull it on another device to restore.
Create a fine-grained PAT with Contents read/write on your backup repo only.
Push after each session to keep a backup accessible from another device.
Save a link to your backup repo here for quick access.
Pushes your full data as readable JSON to a private GitHub repo. No password required.
Your messages will be sent to your AI provider and are subject to their privacy policy. team-sheet never stores or sees your data.
IFS sessions can involve sensitive personal material. Use a provider and model you trust. For full privacy, use a local model via Ollama (no data leaves your device).
Auto-set to cover your data (~... tokens) with room for the conversation. Raise it only if you get context limit errors.
Use full data with models that have 32k+ context. Set context window to at least 2x your data size.
Controls when updated data is written back to the app.
When on, any data update shows a summary of what will change with an Apply button. When off, auto-sync applies silently. Has no effect in Manual mode.
Compaction summarises older messages so the session can continue without hitting context limits.
| RAM | Model | |
|---|---|---|
| 8 GB | gemma3:4b | |
| 16 GB | qwen2.5:7b | |
| 32 GB | qwen2.5:14b | |
| 64 GB+ | qwen2.5:32b |
llmfit detects your RAM and GPU then scores every model. Run it once to get a ranked list tailored to your hardware.
brew install llmfit
1. Download llmfit for Windows and add it to your PATH (or run from the download folder)
1. Download llmfit for Linux and make it executable: chmod +x llmfit && sudo mv llmfit /usr/local/bin/
llmfit fit