@@ -0,0 +1,90 @@
1+<p align="center">
2+ <img src="assets/icon-2-dark-terminal.svg" alt="mini-coder logo" width="112" />
3+</p>
4+
5+<h1 align="center">mini-coder</h1>
6+
7+<p align="center"><strong>Lightning-fast coding agent for your terminal.</strong></p>
8+
9+<p align="center">
10+ <a href="https://www.npmjs.com/package/mini-coder">npm</a>
11+ ·
12+ <a href="https://sacenox.github.io/mini-coder/">docs</a>
13+ ·
14+ <a href="spec.md">spec</a>
15+</p>
16+
17+<p align="center">
18+ <picture>
19+ <img src="assets/preview-0-6-0.png" alt="mini-coder terminal session preview" width="960" style="border-radius: 10px; box-shadow: 0 10px 30px rgba(0,0,0,0.35);" />
20+ </picture>
21+</p>
22+
23+mini-coder (`mc`) is a terminal coding agent, hand crafted for transparency and good engineering performance.
24+
25+## Install
26+
27+```bash
28+$ bun add -g mini-coder
29+$ mc
30+```
31+
32+## Why mini-coder?
33+
34+- **Lean on proven dependencies** — [pi-ai](https://github.com/badlogic/pi-mono/tree/main/packages/ai) for providers, streaming, tool calling, usage tracking, and OAuth. [cel-tui](https://github.com/sacenox/cel-tui) for the terminal UI. The core stays focused on agent work.
35+- **Flat, simple codebase** — no workspaces, no internal abstraction layers. Files grouped by concern in a single `src/` directory.
36+- **Agent-first** — every decision serves the goal of reading code, making changes, and verifying them via the shell.
37+- **Performance** — startup and turn latency matter more than features.
38+- **Streaming end-to-end** — assistant text, reasoning, tool calls, and tool output show up as they happen.
39+
40+## Settings
41+
42+> Warning, settings have changed, update your old settings file!
43+
44+Global defaults live in `~/.config/mini-coder/settings.json`.
45+
46+Create or edit it directly to set the default provider, model, reasoning effort, and any custom models. `customProviders` entries use the pi-ai `Model` shape; the top-level `provider` and `model` select the active entry.
47+
48+```json
49+{
50+ "provider": "ollama",
51+ "model": "llama3.1:8b",
52+ "effort": "medium",
53+ "customProviders": [
54+ {
55+ "id": "llama3.1:8b",
56+ "name": "Llama 3.1 8B (Ollama)",
57+ "api": "openai-completions",
58+ "provider": "ollama",
59+ "baseUrl": "http://localhost:11434/v1",
60+ "reasoning": false,
61+ "input": ["text"],
62+ "cost": {
63+ "input": 0,
64+ "output": 0,
65+ "cacheRead": 0,
66+ "cacheWrite": 0
67+ },
68+ "contextWindow": 128000,
69+ "maxTokens": 32000
70+ }
71+ ]
72+}
73+```
74+
75+Use `api: "openai-completions"` for OpenAI-compatible servers such as Ollama, vLLM, LiteLLM, and local proxies. For local OpenAI-compatible custom providers, mini-coder supplies the dummy API key required by pi-ai when no real key is needed.
76+
77+## Also makes LLMs smarter
78+
79+LLMs famously tell you to walk 50 meters to the car wash — forgetting the car needs to be there too. Not on our watch.
80+
81+<table align="center">
82+ <tr>
83+ <td><img src="assets/mc-claude-smart.png" alt="Claude correctly answering the car wash question" width="400" /></td>
84+ <td><img src="assets/mc-gpt-smart.png" alt="GPT correctly answering the car wash question" width="400" /></td>
85+ </tr>
86+</table>
87+
88+## License
89+
90+MIT