Skip to main content

FAQ

Do I need an API key?​

For the default OpenAI provider, yes — supply OPENAI_API_KEY (or point model.host at an OpenAI-compatible endpoint, where local servers often accept a placeholder key). To run with no key at all, use the fully-local Ollama provider. API providers like Anthropic also need their own key.

Where does AETHON store my data?​

Under ~/.aethon — config (config.yaml), workspace (workspace/), sessions (sessions/), logs (logs/), vector memory (memory.sqlite), and credentials (credentials/).

Is AETHON open source?​

It's source-available under PolyForm Noncommercial 1.0.0 — free for noncommercial use, but not OSI-approved open source (commercial use isn't permitted). See License.

Can I run it fully offline / locally?​

Yes. Install the ollama extra, set provider: ollama, and use Ollama embeddings for memory. No cloud calls are required in that configuration.

How do I expose the Web UI on my network?​

Set channels.webchat.host: 0.0.0.0 and also set dashboard.auth_token. Then reach the dashboard with ?token=YOUR_TOKEN to set the auth cookie.

Which channels need extra installs?​

Only WhatsApp (the whatsapp extra). CLI, WebChat, Telegram, Discord, and Slack all ship in the core install.

How do I add my own workflow?​

Drop a *.sop.md file in ~/.aethon/workspace/sops/ (with an ## Overview section) and invoke it as /<name>. See SOPs.

Will an API provider run up a huge bill?​

Set budget.daily_usd to cap daily spend. Every turn's token usage is measured and costed (override the built-in rate table with budget.pricing); turns are warned once spend crosses budget.warn_ratio (default 0.8) of the ceiling and blocked once it is breached — which also halts ambient and scheduler turns. The default 0.0 means unlimited (measure only). To avoid cloud spend entirely, run fully local with Ollama.

Does the assistant remember things between sessions?​

Yes, when memory is enabled. It stores embeddings in SQLite and retrieves them by similarity. The memory guard prevents secrets from being saved. With memory.auto_recall on (opt-in, off by default), each turn embeds the incoming message and injects the top-matching long-term memories as a prompt layer, so relevant memories surface without the agent calling the memory tool. Vector memory also records each row's embedding model and dimension and refuses to mix dimensions, so changing the embedding model can't silently corrupt similarity search.