OpenQode

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

Add Game Dev Mode screenshot to README

Gemini AI · 2025-12-20 03:47:26 +04:00

33261fdc56

Add macOS and Linux Full-Auto installation commands

Gemini AI · 2025-12-20 02:49:08 +04:00

6c68956c35

Add official screenshot and Full-Auto installation command

Gemini AI · 2025-12-20 02:45:11 +04:00

293fd5d6f0

Release v1.01 Enhanced: Vi Control, TUI Gen5, Core Stability