The page presented WithConverseAPI as a precondition for tool use,
structured output and reasoning, while its own default model is Claude
Haiku 4.5 — and the legacy InvokeModel path builds tools, output_config
schemas and thinking payloads for Claude. Only the non-Anthropic
providers, whose legacy builders never read options.Tools, actually
need Converse.
PAGI-125
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The tutorial's own prerequisites say Go 1.26.5+ and the library's go.mod
requires it, but the Dockerfile stage still pulled golang:1.24-alpine,
so the documented docker build failed at go build with a toolchain
that is older than the module requires.
PAGI-125
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
- Changed Go version requirement from 1.24+ to 1.26.5+ in multiple tutorial and contributing documents to ensure compatibility with the latest features and security updates.
- Updated search index keywords for improved searchability and relevance.
- Set pnpm version to 10.34.5 in the GitHub Actions workflow for publishing documentation.
- Removed pnpm overrides from package.json and added package manager specification.
- Created a new pnpm-workspace.yaml file to manage package overrides for webpack-dev-server.
- Updated search-index.json to refine keyword entries for improved searchability.
- Added new guides for Anthropic and Google Gemini to the getting started section.
- Updated existing guides for Ollama, OpenAI, and Mistral to correct example links.
- Improved clarity in the architecture documentation regarding streaming responses and memory implementation.
- Enhanced structured output handling and reasoning options across various model integrations.
- Fixed minor formatting issues and updated references to ensure consistency throughout the documentation.
- Removed deprecated RedactedThinkingContent from search index and adjusted related tests to enhance clarity.
- Updated CountTokens test to reflect correct expected token count.
- Refactored marshaling tests to streamline reasoning content handling, removing unnecessary redacted content checks.
- Modified CallOptions struct to use pointers for optional fields, improving memory efficiency and consistency.
- Enhanced test cases for CallOptions to validate new pointer-based logic and default values.
- Updated Anthropic client to utilize new CallOptions methods for better encapsulation and maintainability.
- Downgraded React and ReactDOM from version 19.1.0 to 18.3.1 for compatibility.
- Upgraded various Algolia packages, including `@algolia/abtesting`, `@algolia/client-abtesting`, and others, to version 5.49.0 to incorporate the latest features and fixes.
- Updated Docusaurus and related dependencies to version 3.9.2, ensuring alignment with the latest ecosystem changes.
- Adjusted lock files to reflect the updated dependency versions, enhancing overall project stability and performance.
- Introduced reasoning content handling in MessageContent and ToolCall structures, allowing for the preservation of reasoning details during JSON marshaling and unmarshaling.
- Enhanced tests to validate the serialization and deserialization of message content with reasoning, covering various scenarios including text and tool call types.
- Updated marshaling logic to conditionally include reasoning data based on its presence, ensuring accurate representation in JSON format.
- Added new reasoning-related tests to ensure comprehensive coverage of the new functionality.
- Updated reasoning handling in the Anthropic and OpenAI clients to support real-time streaming of reasoning tokens.
- Refactored reasoning structures to improve integration with existing models, ensuring consistent behavior across different LLM implementations.
- Added comprehensive tests for reasoning and caching functionalities, validating the expected behavior in various scenarios.
- Removed deprecated files related to previous caching implementations to streamline the codebase.
This update improves the overall performance and reliability of reasoning capabilities in LLM interactions.
- Added support for recording and replaying multiple identical requests, allowing each to return different responses in a defined order.
- Introduced `replayEntry` struct to manage response cycling for identical requests.
- Updated documentation to reflect new circular buffer behavior and usage examples for cache testing scenarios.
- Enhanced tests to validate the circular response behavior for both GET and POST requests.
This change improves the testing capabilities for scenarios where identical requests yield varying responses, such as caching mechanisms.
* vectorstores/milvus: complete migration to new SDK v2 client (#1397)
Complete the migration from the archived milvus-sdk-go/v2 to the new
github.com/milvus-io/milvus/client/v2 SDK as tracked in issue #1397.
**New Implementation**:
- Add complete vectorstores/milvus/v2/ package with new SDK client
- Implement all core vectorstore operations (Add, Search, Delete, etc.)
- Add comprehensive test suite with unit and integration tests
- Include migration example and documentation
**Key Changes**:
- New milvus.go with updated client initialization and operations
- Updated options.go with v2 SDK configuration patterns
- Added example_migration.go demonstrating upgrade path
- Comprehensive README.md with migration guide
- Updated go.mod/go.sum with new SDK dependencies
**Documentation**:
- Add docs/package-lock.json for documentation build dependencies
- Provide clear migration path from v1 to v2 implementation
- Maintain backward compatibility where possible
This completes the migration work started in earlier commits and provides
a full replacement for the deprecated SDK while maintaining the same
vectorstore interface.
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
* style: apply gofmt formatting to v2 package files
* fix: remove unused async field from v2 Store struct
---------
Co-authored-by: Claude <noreply@anthropic.com>
* docs: fix outdated API references and examples across documentation
Update documentation to reflect current API patterns and fix outdated code examples:
- Update Error struct definition with Provider and Details fields in architecture docs
- Fix code examples to use CallOptions (WithTemperature, WithMaxTokens) instead of constructor options
- Update chain execution pattern from chain.Run() to chains.Run()
- Add context parameter to memory interface methods
- Update Vertex AI examples to use new googleai/vertex package structure
- Change from vertexai.New() to vertex.New() with updated option names
- Fix import paths for Google AI and Vertex AI integration
- Update minimum Go version requirement from 1.21 to 1.23
These changes ensure documentation examples work with the current codebase and provide accurate guidance for developers.
* docs: expand integration documentation and fix remaining API references
- Add comprehensive examples to chat and embeddings integration pages
- Fix missing context imports in LLM provider configuration
- Update Vertex AI examples to use correct vertex.New() syntax
- Fix broken link in CloudSQL vectorstore README
- Transform placeholder pages into useful developer resources
* docs: fix relative link in embeddings integrations page
docs: complete basic chat application tutorial with progressive examples
Completely rewrite the basic chat application tutorial with a structured, step-by-step approach:
**Tutorial Documentation**:
- Restructure tutorial into 6 clear progressive steps
- Add proper setup instructions and prerequisites
- Include code examples for each step with explanations
- Improve clarity and flow from basic to advanced concepts
**Complete Working Example**:
- Add `examples/tutorial-basic-chat-app/` with full implementation
- Include separate files for each tutorial step (step3-step6)
- Add comprehensive README with usage instructions
- Support multiple execution modes via command-line arguments
**Progressive Implementation Steps**:
- Step 3: Basic single-shot LLM interaction
- Step 4: Interactive chat loop without memory
- Step 5: Chat with manual conversation memory management
- Step 6: Advanced chat using chains with automatic memory
**Features Added**:
- Go module setup with proper dependencies
- Error handling and graceful exit functionality
- Multiple chat implementations demonstrating different approaches
- Clear documentation linking tutorial to working code
- Support for running individual steps or complete implementation
The tutorial now provides a complete learning path from basic LLM usage to sophisticated conversation management using LangChainGo's chains and memory systems.
* prompts: add secure template rendering with filesystem support
* prompts: add comprehensive tests and examples
- Add security effectiveness tests for all attack vectors
- Add template rendering tests with all formats
- Add filesystem template tests with MapFS
- Add prompt template examples demonstrating all features
- Improve error handling and add local package references
* docs: add comprehensive prompt template documentation
- Add prompt template user guide with security section
- Add detailed API documentation for new functions
- Update partial values documentation
- Add template-related terms to style guide
- Document migration patterns and security considerations
* prompts: split template implementations by engine type
- Move Go template implementation to templates_go.go
- Move Jinja2 template implementation to templates_jinja2.go
- Clean up imports in templates.go
- Maintain all existing functionality with better code organization
* prompts: move security tests to dedicated file with expanded coverage
- Create new security_test.go file for all template security tests
- Expand test coverage for path traversal attacks in Jinja2 templates
- Add tests for migration patterns from vulnerable to secure implementations
- Remove redundant security tests from templates_test.go
- Improve organization of test code for better maintainability
Sergey Kozyrenko