Skip to content

AI Agent (Eden)

EVE Console ships with an optional built-in conversational assistant — Eden by default (you can rename it). It has read access to the data in your local database via tool calls, so it can answer questions the UI isn't explicitly designed for.

The agent is entirely optional and inactive until you set it up. If you don't configure it, you can ignore it completely. Its settings live in Settings (the ⚙ gear button, top-right) on the AI Agent tab.

What it can do

  • Answer questions about your character/corp data by querying the local database.
  • Run against an external provider (Claude or OpenAI) or a local model (Ollama / LM Studio).
  • Optional text-to-speech so it can talk back, and speech-to-text (with a global push-to-talk key) so you can talk to it hands-free while the game has focus.
  • Customizable name, response verbosity, and voice.

Once enabled, a ✦ {name} button appears in the title bar; click it to toggle the chat panel.

Setup

All settings are on the AI Agent tab. Click Save at the bottom when done.

  1. Personalisation — optionally set an Agent Name (blank = Eden) and a Response Verbosity (Concise, Balanced, or Detailed).
  2. Enable — tick Enable {name} AI companion. This just makes the panel available; you still need a provider configured below.
  3. Provider — choose the LLM Provider, then fill in the section that appears:
    • Claude (Anthropic) — paste an API key (sk-ant-…) and optionally a model (default claude-sonnet-4-6; claude-opus-4-8 and claude-haiku-4-5 also work). Get a key at console.anthropic.com.
    • OpenAI — paste an API key (sk-…) and optionally a model (default gpt-4o). Get a key at platform.openai.com.
    • Local LLM (Ollama / LM Studio) — set the API Endpoint (default http://localhost:11434) and Model Name (default llama3.1). The runner must expose an OpenAI-compatible /v1/chat/completions endpoint, which Ollama and LM Studio do by default.
  4. Context Management — optionally persist chat history to disk across restarts and set a summarization threshold (estimated tokens) at which older messages are silently compacted into a summary. Lower values cut cost per message but drop older context.

Voice (optional)

  • Voice (TTS) — pick a provider to have the agent speak:

    • OpenAI — cloud voices (shares the OpenAI API key above); choose voice, model, and speed.
    • ElevenLabs — cloud voices; needs its own API key and a voice ID.
    • Kokoro — high-quality neural TTS that runs fully offline; the model (~320 MB of neural-network weights, no executable code) downloads on first use.
    • Piper — fast offline neural TTS; the runtime is bundled, and only the chosen voice (a data file) downloads on demand.

    Use Test Voice to preview. Volume and mute are also available in the agent panel itself. - Speech Input (Push-to-Talk) — pick a provider to talk to the agent: - OpenAI Whisper — cloud transcription (shares the OpenAI key); Windows and Linux. - Local Whisper — runs on this machine, no API key; Windows only, and a model file must be downloaded first.

    Then choose a Microphone Device (click ↺ to refresh the list) and a Global Push-to-Talk Key — hold it to record even when EVE has focus. F13–F20 are rarely captured by games and make good PTT keys. The mic button in the panel works regardless of this setting.