THE LOCAL CLI AGENTS
LEGIONHERCULES runs autonomous CLI agents locally through Ollama. The agents aren't chatbots with a tool call bolted on — they plan multi-step tasks, pick the right tools, run them, read the output, and keep going until the job is done. Multiple agents can run in parallel: a coder and a researcher working at the same time, no queue.
Built-in tools cover file read/write/edit, bash, and DuckDuckGo search — all included, no setup. Pick an agent preset (coder for file access and code edits, researcher for web search and summarization, default for everything), switch models with --model codellama / mistral / llama3.1 / phi3, and manage what's installed with legionhercules models --list and --pull.
ollama serveollama pull llama3.2 # drop into an interactive sessionlegionhercules chat # one-shot task with a specific agentlegionhercules chat --agent coder "Fix the bug in utils.py"Want to help?
LEGIONHERCULES needs: Contributions welcome via GitHub — the repo documents dev setup and what's useful right now.. Check CONTRIBUTING.md in the repo, then open a PR — or a bug report with reproduction steps, which is even better.