- Enhanced guessModelInfoFromId method to detect Claude models in AIP ARNs
- Added intelligent prompt caching detection for inference profiles
- Claude-specific profiles get optimized cache settings (4 cache points)
- Generic profiles get conservative cache settings (1 cache point)
- Added comprehensive test suite for AIP prompt caching functionality
- Fixes issue #6429 where AIP ARNs could not use prompt caching
This allows users with Application Inference Profiles to leverage prompt
caching for cost optimization and performance improvements.
feat: remove "(prev Roo Cline)" from extension title in all languages
- Updated all package.nls.*.json files to remove "(prev Roo Cline)" references from extension display names
- Updated all localized README.md files to remove "(prev Roo Cline)" references from titles
- Updated main README.md to remove "(prev Roo Cline)" reference from title
- Affects 18 language files and 18 README files across all supported locales
Co-authored-by: Roo Code <roomote@roocode.com>
feat: increase Claude Code default max output tokens from 8k to 16k
- Changed CLAUDE_CODE_DEFAULT_MAX_OUTPUT_TOKENS from 8000 to 16000
- Users can still lower it to 8k via environment variable if needed
- Addresses issue #6125 regarding output token limits
Co-authored-by: Roo Code <roomote@roocode.com>
- Add todo list initialization and tracking throughout the review workflow
- Replace all MCP GitHub server calls with gh CLI commands
- Fix duplicate step numbering issue
- Update best practices and common mistakes documentation
- Add notes about GitHub CLI limitations for inline comments
* feat: add prompt caching support for LiteLLM (#5791)
- Add litellmUsePromptCache configuration option to provider settings
- Implement cache control headers in LiteLLM handler when enabled
- Add UI checkbox for enabling prompt caching (only shown for supported models)
- Track cache read/write tokens in usage data
- Add comprehensive test for prompt caching functionality
- Reuse existing translation keys for consistency across languages
This allows LiteLLM users to benefit from prompt caching with supported models
like Claude 3.7, reducing costs and improving response times.
* fix: improve LiteLLM prompt caching to work for multi-turn conversations
- Convert system message to structured format with cache_control
- Handle both string and array content types for user messages
- Apply cache_control to content items, not just message level
- Update tests to match new message structure
This ensures prompt caching works correctly for all messages in a conversation,
not just the initial system prompt and first user message.
* fix: resolve TypeScript linter error for cache_control property
Use type assertion to handle cache_control property that's not in OpenAI types
