- Removed the ensureProjectsRoot function and related logic, simplifying project creation to directly create a project without waiting for a root.
- Deleted the DashboardProjectsFolderBar component and its references, as it was no longer needed.
- Updated the createProject function to eliminate the generation of random project names, now defaulting to a simple "Untitled" draft.
- Introduced a new FolderMenu component for managing project folders, including saving functionality.
- Enhanced the ProjectProvider context to handle draft saving and folder management, including keyboard shortcuts for saving.
- Updated documentation to reflect changes in project saving behavior and context reporting.
- Upgraded the @diffusionstudio/api-contract dependency across multiple packages to version 0.5.0.
- Added new model configurations for image, video, and voice categories, including 'nano-banana-2.1', 'flux-3-image', 'minimax-h3-max', and 'elevenlabs-v4'.
- Updated documentation to reflect the new models and their capabilities, ensuring accurate references for users.
- Removed outdated API reference from the `.env.example` file.
- Updated `@diffusionstudio/api-contract` dependency version to `^0.4.0` in both `package.json` files.
- Replaced model name `gemini-3.5-flash` with `qwen3.8-omni-flash` in configuration and documentation, ensuring alignment with the latest model capabilities.
- Added a note in the editor documentation regarding project management in cloud-synced folders to prevent issues with media handling.
- Updated documentation for various media tools to reflect the removal of the `media` prefix from command names, enhancing clarity and consistency.
- Adjusted links and references in the documentation to point to the new command structures for `grab`, `transcribe`, `filmstrip`, `waveform`, and `listen`.
- Added new documentation for the `filmstrip`, `grab`, `listen`, `probe`, and `transcribe` tools, detailing their functionalities and usage.
- Improved error handling descriptions and examples in the documentation to assist users in understanding tool operations better.
- Removed the `media` prefix from various commands (`media_grab`, `media_transcribe`, etc.) to simplify usage and improve consistency across the CLI and API.
- Updated the corresponding handlers and references in the codebase to reflect the new command names.
- Enhanced the README and documentation to clarify the changes in command structure and provide updated examples for users.
- Introduced new handlers for `grab`, `transcribe`, `filmstrip`, `waveform`, and `listen`, ensuring they align with the updated command naming conventions.
- Improved error handling and validation in the new command implementations for better user experience.
- Introduced a new `audioFile` function to extract audio from assets, ensuring only the audio track is uploaded for analysis and transcription.
- Updated `mediaListen` and `mediaTranscribe` handlers to utilize the new audio extraction logic, improving efficiency and clarity in processing audio and video assets.
- Implemented error handling for audio analysis and transcription, ensuring robust responses for unsupported formats and empty audio tracks.
- Added `requireSignIn` and `apiError` functions to manage user authentication and API error handling, enhancing the reliability of tool operations.
- Updated documentation to reflect changes in audio processing and tool functionalities, providing clearer guidance for users.
- Replaced outdated model names in the configuration with new models: `remove-background` and `upscale-image` are now `bria-rmbg-2.0` and `seedvr-2`, respectively.
- Updated the `request` logic in the `use-transforms` file to reflect the new model names for background removal and image/video upscaling.
- Changed the transcription model from `transcribe` to `universal-3.5-pro` in the captions generation process.
- Adjusted documentation to align with the new model names and their functionalities, ensuring clarity for users.
- Added an `--estimate` option to the `generate` command, allowing users to retrieve the estimated credits and execution time without starting a job.
- Updated the `generate` function to handle the new estimate logic, returning a structured response with `credits` and `etaSeconds`.
- Enhanced documentation to clarify the usage of the estimate feature and its implications for job execution.
- Modified related types and validation to accommodate the new estimate functionality, ensuring robust input handling.
- Updated the `generate` function to declare every model's fields by name, allowing clients to send only the declared properties.
- Introduced a new `modelFields` constant that aggregates fields from both API and local models, enhancing input validation.
- Added tests to verify the correct declaration of model fields and ensure proper parsing of input data.
- Eliminated the `media_segment` command and its associated options from the CLI, streamlining the media handling process.
- Removed references to `media_segment` in the API handlers and tests, ensuring a cleaner codebase.
- Updated the `present` function to no longer handle `media_segment`, focusing on other media processing tasks.
- Adjusted related documentation to reflect the removal of the `media_segment` functionality, enhancing clarity for developers.
- Added `generate` and `job` handlers to facilitate media generation and job management through the API.
- Enhanced the `generate` function to handle file uploads and return job details, including estimated run time and credit costs.
- Implemented the `job` function to retrieve job status and results, allowing users to poll for job completion and access generated files.
- Updated documentation to reflect the new endpoints and their usage, improving clarity for developers integrating with the API.
- Removed unused imports and code related to generative assets in various components.
- Updated the `package-lock.json` to remove `@diffusionstudio/api-contract` and `@trpc` dependencies.
- Introduced a placeholder mechanism for generative AI features that are currently being rebuilt.
- Adjusted the CLI and documentation to reflect changes in available commands and features.
- Enhanced the handling of timeouts in the CLI client for media analysis and rendering tasks.
- Updated the CLI to include a new `window` command for managing the application window state (show/hide).
- Improved the background launch logic to prevent launching the app if it is already running in the background.
- Added new tray icons for different application states (busy, idle, window) to enhance user experience.
- Refactored the `agent-chat` module to improve the handling of the agent chat endpoint and its lifecycle.
- Introduced a `WindowHost` class to manage the main window's lifecycle, ensuring it only exists when needed and integrates with the tray functionality.
- Removed the deprecated `headless` module and adjusted related logic to streamline the application flow.
- Updated the DAPI server to support window management, allowing tools to interact with the main window effectively.
- Introduced new options for the `fonts` command to filter by provider and popular fonts, enhancing user experience in font selection.
- Updated the documentation to reflect the inclusion of Google Fonts and local fonts, clarifying how to access and utilize them in projects.
- Implemented a new `FontDropdown` component in the web inspector for improved font selection, allowing users to choose from popular and Google Fonts seamlessly.
- Refactored existing components to integrate the new font selection logic, ensuring consistency across the application.
- Enhanced the `present` function to support background track processing in the `media_segment` tool, allowing for real-time tracking and reporting of mask states.
- Introduced a new `presentContext` function to manage the output of background tracks, including their progress and contact sheets.
- Updated the `context` tool to report the status of ongoing background tracks, including their state and associated images.
- Modified the `media_segment` tool to return tracking states and paths, improving integration with the `<mask>` component in the editor.
- Enhanced documentation to clarify the new behavior and usage of background tracks in the context of media segmentation.
- Introduced the `media_segment` tool to segment objects in video footage using SAM 2.1, allowing users to track objects and generate mask files.
- Implemented input validation for prompts, ensuring at least one point or a bounding box is provided for segmentation.
- Enhanced the output structure to include mask paths and tracking details, improving integration with the `<mask>` component in the editor.
- Added comprehensive documentation for the new tool, detailing usage, input parameters, and expected outputs.
- Updated related components and tests to support the new functionality, ensuring robust performance and user experience.
- Deleted the `whoami` command from the CLI, including its associated action and description.
- Updated various documentation files to remove references to the `whoami` tool, ensuring consistency across the project.
- Adjusted related test cases and API handlers to reflect the removal of the `whoami` functionality.
- Added a new `dev-shim.mjs` script for creating and removing a Windows shim for the `dapi` command, ensuring it points to the correct executable.
- Updated the `package.json` to include new npm scripts for creating and removing the shim.
- Refactored the CLI installation logic to support Windows, including the addition of `refreshCliShim` to rewrite the shim on app updates.
- Enhanced the `mcp-config` to accommodate Windows-specific command handling for the `dapi` binary.
- Introduced tests for the Windows CLI installation and cloud sync functionalities to ensure reliability across platforms.
- Deleted the `fetch` command from the CLI and associated handlers, streamlining the toolset.
- Updated documentation to reflect the removal of the `fetch` command, emphasizing the use of `yt-dlp` directly for downloading videos.
- Adjusted various components and types to remove references to `fetch`, ensuring consistency across the codebase.
- Enhanced error handling to clarify the absence of the `fetch` functionality.
- Introduced a new test suite in `projects.init.test.ts` to validate the behavior of the `initProject` function, ensuring it correctly initializes new projects and handles existing project directories.
- Refactored the `initProject` function in `projects.ts` to streamline project setup, including improved handling of entry files and package.json records.
- Enhanced error handling for invalid paths in the `initProject` function, ensuring only absolute paths are accepted.
- Updated the `scaffold` function to utilize the new `ensureRecord` function, improving code clarity and maintainability.
- Added tests for various scenarios, including creating new projects, preserving existing project data, and handling invalid inputs, to ensure robust project management functionality.
- Updated the `ProjectInfo` documentation to clarify project folder definitions.
- Changed request parameters in `MainRequestMap` for `PROJECTS_LIST` and `PROJECTS_RESOLVE` to use `dirs` and `dir` respectively, enhancing clarity and consistency.
- Refactored the `listProjects` function to accept an array of directories instead of a single root, improving project listing logic.
- Adjusted the `resolveProject` function to streamline project resolution based on directory input.
- Updated various components to reflect changes in project data handling, ensuring compatibility with the new structure.
- Removed the `@modelcontextprotocol/sdk` dependency from `package-lock.json` and `package.json`.
- Updated the CLI client to connect via Streamable HTTP instead of a socket, enhancing communication with the MCP server.
- Simplified the `ping` and `call` functions in `cli-client.ts` to utilize the new connection method.
- Removed the `mcp-proxy.ts` file as it is no longer needed with the new HTTP integration.
- Updated documentation to reflect changes in MCP server access and removed references to the socket transport.
- Cleaned up unused imports and refactored related code for improved clarity and maintainability.
- Updated the `media_transcribe` command to write transcripts to a JSON file, including an output option for specifying the file path.
- Implemented a new `presentTranscript` function to handle the writing of transcripts and return relevant metadata.
- Added filtering options for the `logs` command, including `since` and `contains`, to improve log retrieval capabilities.
- Introduced tests for the new logging features and updated documentation to reflect changes in the `media_transcribe` tool and logging options.
- Revised README.md to clarify the app's MCP server functionality and updated prompt usage instructions, replacing `dapi` references with `diffusion` for consistency.
- Refactored MCP server configuration to use `diffusion` as the server name instead of `dapi`, ensuring alignment with the updated terminology.
- Enhanced tests to reflect changes in server naming and configuration, ensuring accurate validation of the new structure.
- Added new skills and prompts for `editor` and `watch`, improving user guidance and interaction with the app's features.
- Replaced all instances of `runsIn` with `environment` across the DAPI codebase to standardize terminology.
- Updated related tests and tool definitions to reflect the new property name, ensuring alignment in functionality and documentation.
- Enhanced error handling in the DapiServer class to maintain robustness with the updated property references.
- Removed fallback parameters from the `field` function in CLI tools to simplify input handling.
- Updated command options for `media_grab`, `media_filmstrip`, and `media_waveform` to remove default descriptions, relying on schema definitions instead.
- Enhanced the `logs` and `fonts` tools to provide clearer descriptions for input fields.
- Improved the `windowFields` schema to include default value descriptions for better user guidance.
- Adjusted the `models` tool to specify model types in the description, enhancing clarity for users.
- Replaced the `time` argument with `times` in the CLI commands for consistency across tools.
- Updated the `capture` and `media_grab` tools to handle multiple timestamps, improving flexibility in frame extraction.
- Refactored input validation to utilize `zod` for better error handling and user feedback.
- Removed the `help.ts` module as its functionality has been integrated into the main CLI structure.
- Updated documentation to reflect changes in command syntax and input handling, ensuring clarity for users.
- Added `zod` as a dependency in package-lock.json and apps/cli/package.json for input validation.
- Introduced new help.ts module to provide descriptions and validation for CLI commands, ensuring consistency with the app's tool catalog.
- Refactored cli-client.ts to improve timeout handling for tool calls, enhancing error management.
- Created mcp-proxy.ts to facilitate communication between agents and the app's MCP server.
- Updated index.ts to streamline command execution and improve error handling.
- Implemented mcp-config.ts and mcp-install.ts for managing MCP server configurations across different agents.
- Enhanced desktop application to support MCP registration from the menu, improving user experience.
- Added @modelcontextprotocol/sdk as a dependency in package-lock.json and apps/cli/package.json.
- Removed unused @types/ws dependency from package-lock.json and apps/cli/package.json.
- Refactored cli-client.ts to utilize the new SDK for tool interactions, enhancing modularity and performance.
- Updated media handling functions in index.ts to streamline output handling and improve clarity.
- Deleted obsolete protocol.ts and ytdlp.ts files, consolidating functionality within the new SDK.
- Introduced edit-types.ts to define types for source edits, improving type safety and maintainability in the editing process.
- Updated package.json files to include vitest for testing support in both CLI and desktop applications.
- Removed @trpc/client and @trpc/server dependencies from package-lock.json and apps/cli/package.json.
- Updated TypeScript configuration in apps/cli/tsconfig.json to simplify compiler options.
- Refactored cli-client.ts to replace TRPC client calls with direct calls to the DAPI, enhancing modularity and performance.
- Adjusted media handling functions in index.ts to utilize the new call method for better clarity and maintainability.
- Cleaned up protocol.ts by introducing encoding and decoding functions for replies, improving data handling consistency.
- Added @diffusionstudio/dapi as a dependency in package-lock.json, apps/cli/package.json, apps/desktop/package.json, and apps/web/package.json.
- Refactored imports in CLI and desktop source files to utilize types and constants from @diffusionstudio/dapi, enhancing code modularity and maintainability.
- Removed the cli-channels.ts file as its functionality is now encapsulated within the new dapi package.
- Introduced a new MCP server design proposal in docs/mcp-server.md, outlining the transition to a unified API for tool interactions.