docs(llm): Add README

llm/README.md

@@ -0,0 +1,74 @@
+# `llm`
+
+Utilities for text completion.
+
+## Configuration
+
+Currently `llm` sources its config from `$XDG_CONFIG_HOME/voice/config.toml`.
+
+### File structure
+
+```toml
+[transcription]
+model_dir = "/path/to/whisper.cpp/models"
+
+[llm.providers.openai]
+enabled = true
+
+# works well with CLI utilities like `pass` or 1Password's `op`
+api_key_command = [
+  "command",
+  "that",
+  "writes",
+  "api_key",
+  "to",
+  "stdout",
+]
+
+[llm.providers.groq]
+enabled = true
+api_key_command = [ ... ]
+
+[llm.providers.ollama]
+enabled = true
+host = "localhost"
+port = 11434
+```
+
+## Usage
+
+```
+Usage: voice-llm <COMMAND>
+
+Commands:
+  completion
+  list-models
+  help         Print this message or the help of the given subcommand(s)
+
+Options:
+  -h, --help     Print help
+  -V, --version  Print version
+```
+
+### `completion`
+
+```
+Usage: voice-llm completion [OPTIONS] --provider <PROVIDER> <USER_MESSAGE>
+
+Arguments:
+  <USER_MESSAGE>
+
+Options:
+  -p, --provider <PROVIDER>              [possible values: openai, groq, ollama]
+  -s, --system-message <SYSTEM_MESSAGE>
+  -h, --help                             Print help
+```
+
+### `list-models`
+
+```
+Usage: voice-llm list-models
+
+Options:
+  -h, --help  Print help
+```