Update README.md

2026-05-03 15:03:37 +03:00
parent 0bf855f51b
commit 8f8256a85b
1 changed files with 80 additions and 85 deletions
@@ -254,108 +254,103 @@ bind = $mainMod, Q, exec, $capture-ai text
 ## 🔎 ALL APP FEATURES
 <details>
 <summary>For Nerds</summary>
+1. For Nerds
+
 1. Chat management  
-Create new chats (chat_1.json, chat_2.json, …)  
-Switch between chats  
-Delete chat (with confirmation)  
-Rename chat (popover)  
-Pin/unpin chats (📌)  
-Sorting: pinned first, then most recently modified  
+Create, switch, delete, rename, pin/unpin chats. Chats are sorted with pinned chats first, then by last modified time.

-2. Persistent settings (config.json)  
-Remember last opened chat (last_chat)  
-Dark/light theme (dark_mode)  
-Pinned chats (pinned_chats)  
-Model list (ai_models)  
-Per-chat model mapping (chat_models)  
-Store OpenRouter key (open_router_key)  
-Mic mode online/offline (is_mic_online)  
-Online STT model (stt_model_online)  
-Offline whisper.cpp paths (whisper_cpp_bin, whisper_cpp_model)  
+2. Local chat storage  
+All chats are stored locally on the machine under the app cache directory.

-3. Sidebar UI  
-Collapse/expand sidebar (☰)  
-Toggle Chats list  
-Toggle AI Models list  
+3. Section-based chat loading  
+Chats use lazy loading. Only the latest 10 messages are loaded first, and older messages load while scrolling up.

-4. Model management (per chat)  
-Separate active model per chat  
-Selecting a model assigns it to the current chat  
-LRU behavior: recently chosen model moves to top  
-Delete models (cannot delete the last remaining model)  
-If a model is deleted, affected chats fall back to default (self-heal)  
-“AI Models +” to add a model (dialog + OpenRouter models link)  
+4. Persistent configuration  
+Uses one config.json file for settings such as theme, models, pinned chats, last chat, STT, RAG, colors, local providers, and language.

-5. Message selection mode  
-Click to select one or multiple messages  
-Selection counter bar  
-Clear selection (✕)  
-Selected bubbles get an outline  
+5. Per-chat model management  
+Each chat can have its own active AI model. Recently selected models move to the top of the list.

-6. Reference trees (reference chain expansion)  
-Expands used_refs recursively (references of references)  
-Ensures context (including images/refs) isn’t lost  
-Stores:  
-used_refs (indices sent to AI)  
-refs_groups (UI reference preview groups)  
+6. Online and local model support  
+Supports OpenRouter models and local API-based text models such as Ollama.

-7. Reference trees (reference chain expansion)  
-Expands used_refs recursively (references of references)  
-Ensures context (including images/refs) isn’t lost  
-Stores:  
-used_refs (indices sent to AI)  
-refs_groups (UI reference preview groups)  
+7. Local provider settings  
+Local providers can store base URL, startup command, stop command, system prompt, and model parameters.

-8. Regenerate (♻)  
-Appears only when exactly one bubble is selected  
-If user bubble: re-asks the same prompt  
-If bot bubble: finds preceding user prompt + rewrites a similar answer  
-Regenerated prompts are marked (regen) with a distinct style 
-Keeps reference trees during regeneration  
+8. Sidebar UI  
+Collapsible sidebar with separate Chats and AI Models sections.

-9. Copy (📋)  
-Appears only on the single selected bubble  
-If message contains copy ... copy, copies only the inner block  
-Otherwise copies full message  
+9. Context mode switch  
+Each chat can switch between Direct mode and RAG mode.

-10. Code block detection  
-Messages wrapped by copy markers render as a code-style block  
-Code block includes a “Copy” button overlay  
+10. Per-chat conversational RAG  
+Supports short-term memory, summary memory, simple retrieval, and code-aware context.

-11. Image sending / preview  
-Supports a pending image preview before sending  
-Saves image path into chat history  
-Shows a small image preview in message bubbles  
-Clears preview after sending  
+11. Reference tree support  
+Selected references can expand recursively so previous context is not lost.

-12. Typing indicator  
-Animated “Thinking…” while AI is running  
-Removed when response arrives  
+12. Message selection mode  
+Users can select one or multiple messages, clear selection, copy, regenerate, or use them as references.

-13. Scroll behavior  
-Floating scroll-to-bottom button (↓) appears when not at bottom  
-Clicking scrolls to bottom and re-enables auto-scroll  
-Button visibility updates based on scroll position  
-(You also tuned behavior to scroll only on regenerate)  
+13. Regenerate  
+Regenerates from selected user or assistant messages while keeping reference context.

-14. Input shortcuts  
-Enter to send  
-Shift+Enter for newline  
-Added: Ctrl+C / Ctrl+V / Ctrl+X / Ctrl+A  
+14. Copyable code blocks  
+Copy-marked content is rendered as a code-style block with a copy button.

-15. Voice-to-text input  
-🎤 to start recording, ⏹ to stop  
-Uses pw-record if available, otherwise arecord (Linux)  
-After stop, transcribes (offline/online) and appends text into input  
+15. Image generation and image handling  
+Supports image generation, image previews, cached generated images, and image attachments.

-16. Offline STT (whisper.cpp)  
-Uses whisper-cli + ggml-tiny.bin for low CPU transcription  
-Reads .txt output (stdout fallback)  
+16. Document support  
+Supports PDF, DOCX, XLSX, TXT, and MD file creation/output. Generated files are shown with downloadable buttons.

-17. Online STT (OpenRouter)  
-If is_mic_online: true, transcribes via OpenRouter  
-Sends base64 WAV as input_audio  
-Model selected via stt_model_online  
+17. PDF handling  
+If a PDF contains text, it is sent as text content. If it has little/no text, the first pages are converted to PNG images and sent as image_url.
+
+18. Document editing behavior  
+AI can read, analyze, summarize, rewrite, and generate edited document outputs when permission is enabled.
+
+19. File create protocol  
+DOCX, XLSX, PDF, TXT, and MD outputs can be generated from AI responses and returned as downloadable files.
+
+20. XLSX support  
+Can create new XLSX tables and filter existing XLSX files with supported operations.
+
+21. Drag & drop attachments  
+Files and images can be added via drag & drop or file picker.
+
+22. Editable file permission  
+Attached files can be toggled editable. The AI only modifies files when permission is enabled.
+
+23. Voice-to-text input  
+Supports microphone recording and transcription with online or local STT.
+
+24. Offline STT  
+Uses whisper.cpp with configured binary and model paths.
+
+25. Online STT  
+Uses OpenRouter audio-capable models and sends WAV audio as input_audio.
+
+26. Token usage display  
+Shows input, output, and total token usage for each response when enabled.
+
+27. Token price display  
+Can estimate message cost using a configurable token price value.
+
+28. Theme system  
+Supports dark/light theme and custom UI colors.
+
+29. Language system  
+Supports external language files such as Turkish and English, with cached language loading.
+
+30. Prompt chooser  
+Prompt behavior blocks can be enabled/disabled, such as copyable, apply, PDF visual edit, file creation, structured output, and code mode.
+
+31. Terminal control  
+The project also supports terminal usage, excluding STT and some GUI-only features.
+
+32. Linux compatibility  
+Designed for Linux devices, including Arch and Debian/Ubuntu-based systems.
 </details>

 <details>