Merge dev to main for v0.4.4

README.md

@@ -2,7 +2,7 @@
 
 FLASH (**F**ast **L**ocal **A**gent **SH**ell) CLI is an AI-powered command-line interface that allows you to interact with local (or remote) [Ollama](https://ollama.com) models while having the ability to execute shell commands directly or through the AI.
 
-[Watch the video on YouTube](https://www.youtube.com/watch?v=padyQR3tPUs)
+[Watch the original video on YouTube](https://www.youtube.com/watch?v=padyQR3tPUs)
 
 ## Features
 
@@ -133,13 +133,19 @@ python run.py
 ### Internal Commands
 
 - `/help` or `/?`: Display the help message.
-- `/model`: Show the currently active model and Ollama host.
+- `/model`: Pick from the models on this machine, or type a name to
+  download one. `/model <name>` switches straight to one.
 - `/clear`: Clear the conversation history.
 - `/image <path> [prompt]`: Send a local image to the model.
 - `/version`: Show the current version and check GitHub for updates.
 - `/update`: Update Flash to the latest version (requires pipx).
 - `/bye`: Exit the application.
 
+Type `@` anywhere in a message to pick a file out of a dropdown, e.g.
+`why does @flash/theme.py fall back to ASCII?`. Arrow keys and Tab pick
+one, `/` walks into a directory, and the model reads whatever you point
+it at. Dot-entries stay hidden until you type the leading dot.
+
 ### Image Recognition
 
 `/image <path> [prompt]` attaches a local image (`.png`, `.jpg`, `.jpeg`,

docs/CONFIGURATION.md

@@ -46,6 +46,8 @@ Requirements:
 | `MAX_TOOL_ROUNDS` | No | `10` | `1` | Maximum number of tool-calling rounds allowed per request. |
 | `MAX_TOOL_OUTPUT_CHARS` | No | `1200` | `500` | Tool output longer than this is truncated (middle removed) before being sent back to the model. |
 | `MAX_OUTPUT_TOKENS` | No | `1024` | `128` | Maximum tokens the model may generate per response. Maps to Ollama's `num_predict` option. |
+| `NUM_CTX` | No | unset | - | Context window to ask Ollama for: a token count, or `max` for the largest the model's architecture supports. Left unset, the model keeps whatever its Modelfile pins (Flash Onyx pins 65536) and Ollama's default applies to models that pin nothing. The cache is allocated when the model loads, whether or not a session fills it, so raising this costs memory up front. |
+| `SHOW_STATS` | No | `1` | - | Prints a dim line under each reply with the tokens the turn used, how long it took, the generation rate, and how full the context got. `0` hides it. |
 | `VOICE` | No | `0` | - | `1` turns voice mode on at startup: press Enter on an empty prompt to speak, and replies are read aloud. Usually set with `/voice on` rather than by hand. |
 | `VOICE_VOSK_MODEL` | No | `vosk-model-small-en-us-0.15` | - | Name of the [Vosk model](https://alphacephei.com/vosk/models) used for listening. Downloaded to `~/.flash/models` on first use. |
 | `VOICE_PIPER_VOICE` | No | `en_US-amy-medium` | - | Name of the [Piper voice](https://huggingface.co/rhasspy/piper-voices) used for speaking, as `locale-speaker-quality`. |