
iFly-Skills
SkillOfficial iFlytek skills: machine translation, video translation and dubbing, Chinese document proofreading, PDF, image and invoice OCR, speech-to-text and TTS, plus contract review (lawyer check needed). Needs iFlytek open platform services and keys; contract review also needs OCR and LLM APIs.
iFly-Skills 🚀
Official collection of iFLYTEK skills for agent ecosystems and developer workflows.
This repository serves as a centralized skill collection built on top of iFLYTEK AI capabilities. It currently includes reusable skills based on atomic model abilities such as speech, OCR, translation, proofreading, and multimodal understanding.
Beyond atomic skills, this repository is designed to continuously expand toward more scenario-oriented skill packages. In the future, it will include not only single-capability skills, but also workflow-level and solution-level skills for real business scenarios, such as document processing, content production, intelligent review, multimodal analysis, and other integrated task chains.
The goal of this repository is to provide a unified place for skill packaging, versioning, maintenance, and external distribution, making iFLYTEK capabilities easier to understand, install, combine, and evolve in agent-based ecosystems.
⚠️ Beta — This project is under active development. Skills, APIs, and configuration formats may change without notice. We welcome feedback and contributions.
📦 Available Registries
You can browse and install skills from the following registries:
| Registry | URL |
|---|---|
| iFLYTEK Skillhub (Official) | skill.xfyun.cn |
| ClawHub | clawhub.ai/user/iflytek.skills |
| Skills.sh | skills.sh/iflytek/ifly-skills |
| SkillHub (Tencent Cloud) | skillhub.cloud.tencent.com |
| SkillsMP | skillsmp.com/creators/iflytek |
| LobeHub | lobehub.com/skills?q=iflytek |
🌟 Available Skills
Currently, the repository provides the following ready-to-use AI skills:
| Skill Name | Description | Directory |
|---|---|---|
| Hyper TTS | Ultra-realistic text-to-speech synthesis with advanced voice control. | iflytek-hyper-tts |
| Image Understanding | Multimodal capability to analyze and understand image contents. | iflytek-image-understanding |
| Invoice OCR | Specialized OCR for extracting structured data from invoices. | iflytek-ocr-invoice |
| PDF/Image OCR | General optical character recognition for documents and images. | iflytek-pdf-image-ocr |
| Speed Transcription | High-speed audio transcription for voice-to-text scenarios. | iflytek-speed-transcription |
| Text Proofread | Intelligent text proofreading, error detection, and correction. | iflytek-text-proofread |
| Machine Translation | High-quality, multi-language translation capabilities. | iflytek-translate |
| Video Translate | AI-powered video translation with dubbing and multilingual localization. | iflytek-video-translate |
| Voice Clone TTS | Voice cloning technology for customized text-to-speech generation. | iflytek-voiceclone-tts |
| Contract Intelligence Review | Intelligent contract review covering OCR, clause analysis, risk detection, and compliance. | iflytek-contract-intelligence-review |
| Animated Sketch Diagram | Hand-drawn animated architecture diagrams — pure SVG+CSS single-file HTML with seamless-loop GIF export. | animated-sketch-diagram |
🛠️ Usage
Self-hosted n8n
The npm package @iflytekopensource/n8n-nodes-iflytek-skills provides 11 community nodes for speech, OCR, translation, proofreading, image understanding, contract review, and diagram rendering.
In Settings → Community Nodes → Install, enter @iflytekopensource/n8n-nodes-iflytek-skills. The instance must allow unverified community nodes. Prepare the Python environment using the package installation guide, restart n8n, and search for iFlytek in the node selector.
The package requires Node.js 24.x, Python 3.10 or newer, and permission to run local child processes; n8n Cloud is not supported. Remote operations use an iFlytek API credential with authorization and quota for each service. See the package README for requirements, a minimal workflow, and support links.
Individual skills
Each skill is packaged in its own directory containing specific instructions, scripts, and metadata. To use a specific skill:
- Navigate to the target skill's directory (e.g.,
cd skills/iflytek-hyper-tts). - Read the
SKILL.mdfile for detailed API documentation, required environment variables (e.g.,IFLY_APP_ID,IFLY_API_KEY,IFLY_API_SECRET), and usage examples. - Use the provided Python scripts or integrate the capability into your own agent workflow.
🤝 Contributing
We welcome contributions to expand the iFLYTEK skills ecosystem!
- Fork the repository.
- Create your feature branch (
git checkout -b feature/new-skill). - Commit your changes (
git commit -m 'feat: add awesome new skill'). - Push to the branch (
git push origin feature/new-skill). - Open a Pull Request.
Note: When you submit a Pull Request, our CLA Assistant bot will automatically ask you to sign the Contributor License Agreement (CLA). Please follow the bot's instructions to complete the PR process.
To avoid unrelated signatures/version1/cla.json diffs and merge conflicts in fork-based Pull Requests, CLA signatures are stored in a dedicated remote signature repository instead of this codebase.
Maintainer setup: configure repository variables CLA_SIGNATURE_STORE_ORG, CLA_SIGNATURE_STORE_REPO, and optional CLA_SIGNATURE_STORE_BRANCH, plus repository secret CLA_SIGNATURE_STORE_PAT with write access to the remote signature repository. The branch used to store signatures should not be protected.
📄 License
This project is licensed under the Apache License 2.0.
💬 Community
Join the Astron Open Source Community (WeCom Group) to discuss and collaborate:

関連リソース

iFlytek Translate
SkillOfficial iFlytek skill that translates text or text files via iFlytek's machine translation API across Chinese, English, Japanese, Korean, French, Spanish, Arabic and more, up to 4,096 bytes per call. Needs Python 3 and iFlytek open platform credentials; paid beyond the free quota.
iflytek/iFly-Skills212
iFlytek Video Translate
SkillOfficial iFlytek skill (beta): translates and AI-dubs videos between Chinese, English, Japanese, Korean, Russian, French, Spanish, Arabic and Cantonese, tracks jobs, fetches subtitles and video. Needs iFlytek API keys and Python 3; input must be an online MP4 (Alibaba OSS per docs), up to 5 min.
iflytek/iFly-Skills212
visa-doc-translate
SkillOCRs photos of visa documents such as bank, employment or income certificates, translates them into English and builds a side-by-side bilingual PDF. Needs Python (pillow, reportlab) and an OCR engine. Not an official certified translation: verify names, amounts and dates.
affaan-m/ECC281k
NoizAI Skills
SkillVoice skills: TTS with subtitle timing, emotional character voices, translated dubbing, two-host news podcasts, sound effects and songs. Narration and podcasts work with local Kokoro or free voices; emotion, cloning, dubbing, effects and songs need a Noiz API key. Never clone voices to deceive.
NoizAI/skills529
DeepL CLI
CLIDeepL's official CLI: translate text and PDF, Word, PowerPoint and Excel documents while keeping formatting, with glossaries and translation memory. Needs a DeepL API key (free plan available); install from source with Node.js 24.15+. Have legal translations reviewed.
DeepL/deepl-cli46
Buzz
CLIOffline speech transcription and translation based on Whisper; audio never leaves your machine. buzz add exports lectures, interviews and meeting recordings to TXT, SRT or VTT. pip install needs Python 3.12 and FFmpeg; a desktop app is also available. Check transcripts manually.
chidiwilliams/buzz22k