Vidu official video generation. Image to video / Text to video / Reference to video / Text to image / Reference to image / Video edit / Image edit

vidu-skill 1.4.2 - Added video composition support via `vidu-cli task compose`, enabling timeline-based video creation (see references/compose.md for required schema). - New commands for querying credit cost, daily quota (claw-pass), and user credit balance. - Expanded main capabilities: now includes video-compose, task cost estimation, and quota queries alongside generation features. - Documentation additions: added English README, updated usage instructions, and included references/compose.md for compose task guidance. - Enhanced guidance for `schedule-mode` auto-detection and model version options.

- Switched CLI dependency from Rust/cargo (`vidu-cli` ≥0.2.0) to Node.js/npm (`vidu-cli@latest`); now requires Node.js ≥14. - Added new capabilities: lip-sync (口型同步), text-to-speech (文字转语音), and related commands. - Updated command references and output contract to support new CLI command set, including audio/video download with `--output`. - Adjusted setup and fallback instructions for npm-based installation. - Updated compatibility and dependency information accordingly.

- Initial release of vidu-skills (version 1.0.5). - Enables video and image generation via the official Vidu API using the vidu CLI. - Supports multiple workflows: text-to-image, text-to-video, image-to-video, reference-based generation, head-tail frame video creation, and reference material management. - Requires VIDU_TOKEN and vidu-cli ≥ 0.2.0; optionally supports custom API base URL. - Provides robust error handling and clear CLI output for both task submission and task monitoring. - Includes bundled reference documentation for parameter options, error handling, and usage guidance.

vidu-skill 1.0.4 - Added Q3 (model_version 3.2) support for reference-to-video (参考生视频) tasks; these now allow Q3 or Q2 models. - Updated documentation to clarify that for reference-to-video, Q3 supports durations 1–16s and Q2 supports 2–8s. - Adjusted transition requirements: for reference-to-video (参考生视频), transition must not be sent (applies to both Q2 and Q3). - No code or functional changes; this is a documentation update for task compatibility and parameter usage.

Initial public release of vidu-skills — Vidu API integration for video/image generation. - Provides endpoints for text-to-image, text-to-video, image-to-video, head-tail-image-to-video, reference-based generation, and material/reference creation. - Requires VIDU_TOKEN (and optional VIDU_BASE_URL) for authentication. - Uses curl (or HTTP client) for direct API calls: including upload, task submission, querying, and SSE-based state streaming. - Includes privacy and security usage notes; details supported task types, input constraints, and configuration. - Supports resolutions up to 4K; workflow instructions and API endpoints detailed in SKILL.md.

v1.0.1 — Major change: switched from local script-based execution to direct API integration using curl. - Removed all bundled Python scripts; no external script execution required. - All functionality (video creation, upload, querying, material management) now uses official Vidu API endpoints directly with curl or any HTTP client. - Skill now requires only VIDU_TOKEN (and optional VIDU_BASE_URL) in the environment. - Updated documentation and examples to describe direct API usage and streamlined setup. - No Python or additional dependencies needed—just HTTP requests.

vidu-skill 1.0.0 - Initial public release of vidu-video-generation skill. - Enables video and material (element) creation via bundled Python scripts—text2video, img2video, headtailimg2video, character2video, and element management. - All interactions use provided scripts; direct API calls are not implemented by this skill. - Supports environment configuration with VIDU_TOKEN and optional VIDU_BASE_URL. - Includes detailed documentation of supported video generation types, arguments, and workflows for both China and International users.