mirror of
https://github.com/LibreChat-AI/librechat.ai.git
synced 2026-03-27 10:48:32 +07:00
9e1a0db393
* chore: update GitHub Actions workflow to use latest action version for improved stability * chore: update roadmap references and enhance documentation for AWS Bedrock inference profiles - Updated footer menu and card icons to reflect the 2026 roadmap. - Adjusted default values in changelog and configuration documentation for `maxRecursionLimit`. - Added comprehensive documentation for AWS Bedrock inference profiles, including setup, configuration, and examples. - Introduced Agents API documentation for programmatic access to LibreChat agents. - Enhanced existing documentation for clarity and consistency across various sections. * feat: release Config v1.3.4 with new features and updates - Introduced `create` field in `interface.prompts` and `interface.agents` for enhanced user control. - Added `interface.remoteAgents` configuration for managing remote agent permissions. - Updated `endpoints.bedrock` with `models` and `inferenceProfiles` for better customization. - Included Moonshot as a known endpoint for AI integration. - Introduced new agent capabilities: `deferred_tools` and `programmatic_tools`. - Removed deprecated `forcePrompt` setting from configurations. - Updated default model lists and added support for new models. - Enhanced `modelSpecs` with `artifacts` field and `effort` parameter for Anthropic models. * refactor: update BlogHeader to use usePathname for route handling - Replaced useRouter with usePathname for improved routing in BlogHeader component. - Simplified page retrieval logic by directly using pathname for matching routes. * feat: add changelog for v0.8.3-rc1 release with new features and fixes - Introduced several enhancements including event-driven lazy tool loading, token usage tracking, and programmatic tool calling UI. - Added support for new models and providers, including Claude Opus 4.6 and Moonshot. - Implemented various bug fixes and improvements for better user experience and performance. * chore: nextjs artifact * first draft roadmap * feat: enhance BlogPage with Open Graph image support and layout improvements - Added support for Open Graph images in blog entries to improve visual presentation. - Refactored article layout for better structure and readability, including adjustments to the display of metadata and content. - Updated styles for improved user experience during hover interactions. * feat: enhance BlogPage with date formatting and layout adjustments - Added a new dateFormatted field to entries for improved date display. - Implemented a date formatter for consistent date presentation. - Refactored article layout to use a grid system for better responsiveness. - Updated styles for article links and metadata for enhanced user experience. * feat: add responsive image sizes to BlogPage for improved layout - Included sizes attribute for Open Graph images to enhance responsiveness on different screen sizes. * feat: update feature titles and descriptions for clarity - Changed titles for "Forking Messages and Conversations" to "Forking Chats" and "Memory" to "User Memory" for better alignment with functionality. - Updated descriptions for "Message Search" and "Upload as Text" to enhance understanding of features. * chore: update configuration version to 1.3.4 across multiple documentation files - Updated the version number in `librechat.yaml` examples to reflect the latest release (1.3.4) in various configuration and feature documentation files. * feat: enhance User Memory documentation for clarity and detail - Updated the description to clarify that User Memory is a key/value store that operates on every chat request. - Added a callout to distinguish between key/value storage and conversation memory. - Expanded on the functionality of the memory agent, including its execution process and user control features. - Introduced a section on future improvements for the memory agent's efficiency and relevance. * feat: update title and description for NGINX documentation - Changed the title from "Secure Deployment with Nginx" to "NGINX" for brevity. - Updated the description to provide a clearer overview of the guide's purpose in securing LibreChat deployment with Nginx as a reverse proxy and HTTPS. * feat: update 2026 roadmap with key accomplishments and future plans - Celebrated LibreChat's 3rd anniversary with a summary of achievements from 2025, including growth in GitHub stars and community engagement. - Clarified the timeline for open-sourcing the Code Interpreter API by the end of Q1. - Revised notes on the v1 Admin Panel's core capabilities and community-driven items for better clarity and detail. * feat: enhance blog and author components with Open Graph image support - Added optional `ogImagePosition` field to blog entries for better image placement control. - Updated BlogPage and individual post pages to utilize the new `ogImagePosition` for responsive image styling. - Improved Author component to conditionally render author images based on availability. - Updated 2026 roadmap blog post with a new Open Graph image and position for enhanced visual appeal. * feat: enhance CardComponent with icon support and layout improvements - Added optional `icon` prop to CardComponent for better visual representation. - Updated CardComponent layout to include icon alongside title and children. - Improved styling for CardComponent and CardsBase for enhanced responsiveness and user experience. * feat: update 2026 roadmap with detailed focus areas and community-driven items - Added sections for Q1 and Q2 focus areas, outlining major initiatives like Dynamic Context and Admin Panel. - Enhanced clarity on community-driven items and their prioritization based on GitHub reactions. - Included hiring information to attract full-stack developers for ongoing project support. - Improved overall structure and readability of the roadmap content. * fix: improve icon styling in CardCompat component for better responsiveness - Updated icon container styling to ensure consistent height and width for SVG icons. - Enhanced layout of CardCompat to maintain visual integrity across different screen sizes. * chore: update .gitignore to include next-env.d.ts for TypeScript support * fix: correct import statement formatting in next-env.d.ts for consistency * fix: refine wording in 2026 roadmap for clarity - Updated the description of agentic workflows to emphasize a lean approach to context pulling. - Enhanced overall readability of the section on Dynamic Context. * feat: expand Admin Panel section in 2026 roadmap with detailed capabilities - Added comprehensive descriptions of the Admin Panel's core functionalities, including GUI for configuration, configuration profiles, group and role management, and access controls. - Clarified the development approach for the Admin Panel, emphasizing ongoing iteration and community involvement. - Updated note on the Admin Panel's prioritization and requirements following the ClickHouse acquisition. * feat: add TrackedLink component for enhanced analytics tracking - Introduced a new TrackedLink component that integrates Vercel analytics to track user interactions with links. - The component allows for customizable link properties while ensuring tracking of clicks with relevant metadata. - Updated CardCompat to utilize the new TrackedLink for improved user engagement tracking. * feat: enhance blog post layout and introduce TrackedAnchor component for link tracking - Wrapped the InlineTOC component in a div for improved spacing in blog posts. - Added a new TrackedAnchor component to facilitate link tracking with Vercel analytics, allowing for customizable anchor elements. - Updated mdx-components to utilize TrackedAnchor for enhanced link interaction tracking. * feat: update TrackedLink and TrackedAnchor components for external link handling - Enhanced the TrackedLink component to differentiate between internal and external links, using Next.js Link for internal navigation. - Introduced a utility function to determine if a link is external, improving tracking accuracy. - Updated TrackedAnchor to utilize the same external link handling logic for consistency in link tracking. * feat: add uncaught exception handling section to dotenv configuration documentation - Introduced a new section on uncaught exception handling, explaining how to override the default behavior to keep the app running after exceptions. - Added an option table detailing the `CONTINUE_ON_UNCAUGHT_EXCEPTION` configuration. - Included a warning callout advising against using this feature in production environments. * feat: add ESLint rule for unused variables in TypeScript - Introduced a new ESLint rule to enforce the handling of unused variables, allowing for specific patterns to be ignored. - This enhancement aims to improve code quality by ensuring that developers are alerted to potentially unnecessary variables while maintaining flexibility in naming conventions. * fix: update copyright year in LICENSE file to 2026 * feat: update footer menu link and add 2026 roadmap blog post - Changed the roadmap link in the FooterMenu component to point to the new blog post. - Introduced a new blog post detailing the 2026 roadmap for LibreChat, outlining key features and focus areas for the upcoming year. - Updated the import statement in next-env.d.ts for consistency with the new types directory. * fix: update import path in next-env.d.ts and add comment block in agents.mdx - Changed the import statement in next-env.d.ts to reference the new development types directory. - Added a comment block in agents.mdx to indicate that the Programmatic Tool Calling feature is in private beta. * fix: remove unused ESLint disable comment in context.tsx * chore: update blog
54 lines
3.4 KiB
Plaintext
54 lines
3.4 KiB
Plaintext
---
|
|
date: 2026-02-18
|
|
title: ⚙️ Config v1.3.4
|
|
version: "1.3.4"
|
|
---
|
|
|
|
- Added `create` field to `interface.prompts` and `interface.agents` configuration
|
|
- Controls whether users can create new prompts/agents (default: `true`)
|
|
- Complements existing `use`, `share`, and `public` permission fields
|
|
- Boolean values still supported for backward compatibility
|
|
- See [Interface Configuration](/docs/configuration/librechat_yaml/object_structure/interface#prompts) for details
|
|
|
|
- Added `interface.remoteAgents` configuration for remote agent access control
|
|
- Added `use` setting to enable/disable remote agent usage (default: `false`)
|
|
- Added `create` setting to control whether users can create remote agents (default: `false`)
|
|
- Added `share` setting to control whether users can share remote agents (default: `false`)
|
|
- Added `public` setting to enable public sharing of remote agents (default: `false`)
|
|
- See [Interface Configuration](/docs/configuration/librechat_yaml/object_structure/interface#remoteagents) for details
|
|
|
|
- Added `models` and `inferenceProfiles` to `endpoints.bedrock` configuration
|
|
- `models`: Specify custom model IDs available for Bedrock
|
|
- `inferenceProfiles`: Map inference profile IDs to display names for cross-region inference
|
|
- See [AWS Bedrock Configuration](/docs/configuration/librechat_yaml/object_structure/aws_bedrock#inferenceprofiles) for details
|
|
|
|
- Added Moonshot as a known endpoint
|
|
- New `moonshot` known endpoint for Moonshot AI integration with icon support
|
|
- See [Moonshot Configuration](/docs/configuration/librechat_yaml/ai_endpoints/moonshot) for details
|
|
|
|
- Added `deferred_tools` and `programmatic_tools` agent capabilities
|
|
- `deferred_tools`: Enables lazy/event-driven loading of MCP tools, reducing context window usage (enabled by default)
|
|
- `programmatic_tools`: Enables programmatic tool calling for MCP tools via code execution sandbox (disabled by default, requires latest Code Interpreter API)
|
|
- See [Agents Configuration](/docs/configuration/librechat_yaml/object_structure/agents) for details
|
|
|
|
- Removed deprecated `forcePrompt` setting
|
|
- Removed from Azure endpoint and custom endpoint configurations
|
|
- Setting was deprecated and no longer functional
|
|
|
|
- Updated default model lists
|
|
- Added `claude-sonnet-4-6` and `claude-opus-4-6` to Anthropic and Bedrock default models
|
|
- Added `moonshotai.kimi` and `moonshotai.kimi-k2.5` to Bedrock default models
|
|
- Added Bedrock parameter support for MoonshotAI, Z.AI, and OpenAI GPT-OSS models
|
|
- Removed deprecated Claude models: claude-3-opus, claude-3-sonnet, claude-3-haiku, claude-2.x, claude-1.x, and claude-instant variants
|
|
|
|
- Added `artifacts` field to `modelSpecs` list entries
|
|
- Enables the Artifacts capability for a model spec
|
|
- Set to `true` to enable with default mode, or a mode string (e.g., `"default"`) for a specific mode
|
|
- See [Model Specs Configuration](/docs/configuration/librechat_yaml/object_structure/model_specs#artifacts) for details
|
|
|
|
- Added `effort` parameter for Anthropic Adaptive Thinking in `modelSpecs` presets
|
|
- Controls the thinking effort level for supported Anthropic models (e.g., Claude Opus 4.6)
|
|
- Options: `"low"`, `"medium"`, `"high"`, `"max"`, or unset (model decides)
|
|
- Supported by `anthropic` and `bedrock` (Anthropic models) endpoints
|
|
- See [Model Specs Configuration](/docs/configuration/librechat_yaml/object_structure/model_specs#effort) for details
|