Hermes Atlas
Automations & use cases

Loki Agent

wundercorp/loki

Self-improving agent with a TUI, messaging gateway and learning loop, tagged for Hermes Agent

In short

Loki Agent is a self-improving AI agent from WunderCorp with a terminal interface, a multi-platform messaging gateway, a learning loop and seven terminal backends. It is tagged hermes-agent and nous-research on GitHub.

What Loki Agent does

Loki Agent is a self-improving AI agent from WunderCorp, Inc. The README describes a terminal interface with multiline editing, slash-command autocomplete and streaming tool output, and a single gateway process that serves Telegram, Discord, Slack, WhatsApp, Signal and the CLI. Its learning loop creates skills from experience, improves them during use, searches past conversations with FTS5 and builds a user model across sessions, and skills follow the agentskills.io open standard.

It includes a built-in cron scheduler with delivery to any platform, isolated subagents for parallel work, and seven terminal backends: local, Docker, SSH, Singularity, Modal, Daytona and Vercel Sandbox. Models are switched with loki model across providers such as OpenRouter, OpenAI or your own endpoint. The installer sets up uv, Python 3.11, Node.js, ripgrep and ffmpeg, plus a portable Git Bash on native Windows. The GitHub topics include hermes-agent and nous-research, but the opening part of the README does not describe how the project relates to Hermes Agent.

Key features

  • Terminal interface with multiline editing, slash-command autocomplete and interrupt-and-redirect
  • One gateway process for Telegram, Discord, Slack, WhatsApp, Signal and the CLI
  • Learning loop that creates and improves skills, with FTS5 session search
  • Built-in cron scheduler with delivery to any platform
  • Seven terminal backends: local, Docker, SSH, Singularity, Modal, Daytona and Vercel Sandbox
  • Model switching with loki model across OpenRouter, OpenAI and custom endpoints

When to use it

  • Running a persistent agent on a cheap VPS and chatting with it from Telegram
  • Scheduling daily reports or nightly backups in natural language
  • Switching model providers without code changes

Who it is for: Developers who want a self-hosted, self-improving terminal agent with messaging access and a Hermes-style feature set.

How it fits with Hermes Agent

Tagged hermes-agent and nous-research on GitHub, and its feature list closely parallels Hermes Agent's, but the opening part of the README does not state its exact relationship to Hermes.

How to install Loki Agent

These commands are copied from the project's README. Check the repository for the latest steps before you run them.

curl -fsSL https://loki.computer/install.sh | bash
iex (irm https://loki.computer/install.ps1)

Requirements: A model provider API key (first-run setup defaults to OpenRouter); the installer adds uv, Python 3.11, Node.js, ripgrep and ffmpeg

Note: The opening part of the README does not explain how the project relates to Hermes Agent, although it is tagged hermes-agent and nous-research.

FAQ

What is Loki Agent?

Loki Agent is a self-improving AI agent from WunderCorp. It combines a terminal interface, a messaging gateway, scheduled automations, subagents and a learning loop that builds skills from experience.

Does Loki Agent work with Hermes Agent?

The opening part of the README does not say. The repository is tagged hermes-agent and nous-research, and its features and skills standard match Hermes Agent's, but check the project directly for its exact relationship.

How do I install Loki Agent?

On Linux, macOS, WSL2 or Termux run curl -fsSL https://loki.computer/install.sh | bash. On native Windows, run iex (irm https://loki.computer/install.ps1) in PowerShell.

Similar automations for Hermes Agent

All automations

Related guides: Build agent teams with the Hermes kanban board · How to run Hermes Agent securely