10 Commits

• fix(automation): Switch to Build Mode for Backend Generation

  - Modified Views.tsx to keep the user in GlobalMode.Build instead of switching to Chat when generating a backend.
  - Added intelligent system prompt routing in LayoutComponents.tsx for [BACKEND_REQUEST] to ensure the AI knows how to generate a server.js file.

  Gemini AI · 2025-12-20 16:05:22 +04:00
  c1b5adbde8
• feat: Add backend generation capability

  - Added 'Build Backend' button to Preview toolbar in Views.tsx
  - Defined BACKEND_GENERATOR_PROMPT in automationService.ts
  - Implemented specific prompt logic to infer API endpoints from frontend code

  Gemini AI · 2025-12-20 15:59:56 +04:00
  71c27fb060
• feat: Add Retry button to chat plans and Cancel AI button to build screen

  - Added a 'Retry' button to plan messages in chat, allowing one-click rejection and regeneration.
  - Added a 'Cancel AI' button to the Building screen to abort stuck or unwanted build processes.

  Gemini AI · 2025-12-20 15:52:29 +04:00
  5954b091c0
• fix: Robust patch generation and safety timeouts

  - Condensed Canvas Isolation Architecture to streamline AI focus
  - Added robust JSON extraction (brace-finding) to handle conversational AI output
  - Implemented 90s safety timeout for patch generation to prevent infinite 'Generating' state
  - Throttled status updates to improve UI performance during streaming

  Gemini AI · 2025-12-20 15:03:13 +04:00
  7785feed1d
• feat: Implement Canvas Isolation Architecture for surgical code precision

  - Replaced old refinement protocol with technical Canvas Isolation Architecture
  - Integrated layered canvas logic (Base, Change, Merge, Validation)
  - Enforced isolation chamber and quarantine zone constraints in prompts
  - Updated QA self-check gate to strictly monitor for scope creep

  Gemini AI · 2025-12-20 14:43:03 +04:00
  0c76288d01
• fix: Enable Ollama models for code generation

  - Added getActiveModel() function to automationService for dynamic model selection
  - Replaced all hardcoded 'qwen-coder-plus' strings with getActiveModel() calls
  - Added localStorage sync when model is changed in orchestrator
  - This enables Ollama and other models to work in all automation tasks

  Gemini AI · 2025-12-20 13:56:21 +04:00
  4230159036
• fix: Ollama click handler and stream completion

  - Fixed sidebar Ollama Cloud section to always open modal on click
  - Improved stream handling with proper buffer flush on end
  - Added 120s timeout for Ollama requests
  - Added better logging for debugging
  - Fixed activeRequest cleanup on errors

  Gemini AI · 2025-12-20 13:29:30 +04:00
  20aef0fd89
• fix: Fix Ollama API key storage - keytar ESM import and file fallback

  - Fixed keytar ESM import to use .default property
  - Added comprehensive error handling and logging for vault operations
  - Fixed fallback to file-based storage when keytar is unavailable
  - Added debug logs to help troubleshoot key storage issues

  Gemini AI · 2025-12-20 13:07:09 +04:00
  94ae6fa8c5
• feat: Add Ollama Cloud integration with 20+ free AI models

  - Added AI Model Manager to sidebar for quick model switching
  - Integrated Ollama Cloud API with official models from ollama.com
  - Added AISettingsModal with searchable model catalog
  - Models include: GPT-OSS 120B, DeepSeek V3.2, Gemini 3 Pro, Qwen3 Coder, etc.
  - Added 'Get Key' button linking to ollama.com/settings/keys
  - Updated README with Ollama Cloud documentation and free API key instructions
  - Fixed ChatPanel export issue
  - Added Brain icon for reasoning models

  Gemini AI · 2025-12-20 13:01:52 +04:00
  e52098f3a8
• Release v1.01 Enhanced: Vi Control, TUI Gen5, Core Stability
  Gemini AI · 2025-12-20 01:12:45 +04:00
  142aaeee1e