Skip to content

quickstart.md

Latest commit

 

History

History
92 lines (64 loc) · 3 KB

quickstart.md

File metadata and controls

92 lines (64 loc) · 3 KB

Quickstart

Get a running agent in 3 commands.

Prerequisites

3-Command Setup

pip install hexis
hexis init --character hexis --provider openai-codex --model gpt-5.2
hexis chat

This uses ChatGPT Plus/Pro OAuth (no API key needed). hexis init opens a browser window for login, starts Docker, pulls the embedding model, configures the character, and runs consent -- all in one command.

Other Providers

OAuth (no API key needed):

# GitHub Copilot (device code login)
hexis init --character jarvis --provider github-copilot --model gpt-4o

# Chutes (free inference)
hexis init --character hexis --provider chutes --model deepseek-ai/DeepSeek-V3-0324

# Google Gemini CLI
hexis init --provider google-gemini-cli --model gemini-2.5-flash --character hexis

# Qwen Portal
hexis init --provider qwen-portal --model qwen-max-latest --character hexis

API-key providers:

# OpenAI Platform (auto-detect provider from key prefix)
hexis init --character jarvis --api-key sk-...

# Anthropic
hexis init --provider anthropic --model claude-sonnet-4-20250514 --api-key sk-ant-...

# Ollama (fully local, no API key needed)
hexis init --provider ollama --model llama3.1 --character hexis

# Express defaults (no character card)
hexis init --api-key sk-ant-...

hexis init auto-detects the provider from API key prefixes. For all supported providers, see Auth Providers.

Verify It Worked

hexis status    # shows agent status, memory counts, energy level
hexis doctor    # checks Docker, DB, embedding service health

Enable Autonomy (Optional)

hexis up --profile active

With the active profile, the agent wakes on its own, reviews goals, reflects, and reaches out when it has something to say. Without it, the agent only responds when you talk to it.

What Just Happened

  1. hexis init started a PostgreSQL container (the agent's brain), pulled an embedding model into Ollama, configured your chosen character's identity/personality/values, and ran a consent flow where the agent agreed to begin.
  2. hexis chat opened an interactive conversation with memory enrichment -- your messages are augmented with relevant memories, and the agent forms new memories from the conversation.

Next Steps