Stilla

A CLI companion for the Great Work of self-development. Stilla is an AI conversational partner that is patient, honest, reflective, and grounded in seven living disciplines — the Seven Pillars.

All your data stays local. Self-model files, journal entries, and configuration are plain Markdown stored in ~/.stilla/, readable and editable with any text editor.

Prerequisites

• Node.js >= 20 (LTS)
• An Anthropic API key

Installation

npm install -g @mhaglind/stilla

Configuration

Set your API key as an environment variable. Add this to your shell profile (~/.zshrc, ~/.bashrc, etc.) to persist it:

export ANTHROPIC_API_KEY="sk-ant-..."

Optionally override the default model:

export STILLA_MODEL="claude-opus-4-20250514"

Or create ~/.stilla/config.md:

---
created: 2026-03-17
last_modified: 2026-03-17
---

# Stilla Configuration

## Model

model: claude-sonnet-4-20250514

Usage

Start a conversation

stilla

Type a message and Stilla responds in character, streaming token-by-token. The conversation maintains full context within the session. Type exit, quit, or press Ctrl+C to end.

View your self-model

stilla self-model

Displays the contents of your self-model files: identity, patterns, relationships, growth, and seasons. These are plain Markdown files in ~/.stilla/self-model/ that you can edit directly with any text editor.

View your journal

stilla journal

Shows the most recent journal entry. Journal entries are reflective summaries generated by Stilla at the end of each conversation, saved to ~/.stilla/journal/.

What happens on first run

1. Stilla creates ~/.stilla/ with self-model templates and an empty journal directory
2. Self-model templates contain reflective guiding questions to help you begin
3. Existing files are never overwritten

Data storage

~/.stilla/
├── self-model/
│   ├── identity.md
│   ├── patterns.md
│   ├── relationships.md
│   ├── growth.md
│   └── seasons.md
├── journal/
│   └── 2026-03-17.md
└── config.md              (optional)

License

Private.