mini-me — documentation (English)
mm is a coding agent that runs in your terminal, or inside an editor through the Agent Client Protocol. It talks to a local model server (Docker Model Runner or llama.cpp's llama-server) and works with a handful of built-in tools: bash, read_skill, read_file, write_file, edit_file.
This documentation is organised along the four Diátaxis quadrants. Pick the one that matches what you need right now.
Tutorials — learn by doing
How-to guides — get a task done
- Run mini-me on llama.cpp instead of Docker Model Runner
- Use mini-me from an editor (ACP)
- Add a skill
- Keep a long session inside the context window
- Start a new session without restarting
- Point the model at a file with
@path
Reference — look it up
- Command line — flags, environment variables, start-up banner
- Slash commands —
/quit,/abort,/compact,/new, keyboard shortcuts - Configuration file — every YAML key, its default and its validation rule
- Built-in tools —
bash,read_skill,read_file,write_file,edit_file - Skill file format
- Agent Client Protocol surface
Explanation — understand
- Architecture: one loop, two front ends
- Providers: one implementation per protocol
- Context compression
- Package diagram (draw.io source):
docs/diagrams/packages.drawio
Scope
This documentation covers the mm binary and the packages under internal/. The kits/ and tmp/ directories at the repository root are not part of the project and are not documented here.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 |
|