AI & Machine Learning skills

Provides OpenAI guidance and assistance for software development workflows.

|

>

>-

Helps with Excel/spreadsheets, Anthropic as part of agent workflows workflows.

AI desktop agent — control any app on Windows/macOS from your OpenClaw agent. Send natural language tasks and it handles opening apps, clicking buttons, typing text, and navigating browsers.

Semantic search sidecar for Karakeep bookmarks. Find bookmarks by meaning, not just keywords. Uses vector embeddings for intent-aware retrieval.

Local speech-to-text using faster-whisper — 4-6x faster than OpenAI Whisper with identical accuracy. GPU acceleration, SRT/VTT/TTML subtitles, speaker diarization, and chapter detection.

Async deep research via Gemini Interactions API. RAG-ground queries on local files, preview costs with --dry-run, get structured JSON output. Universal skill for 30+ AI agents.

Replicates voices from audio samples to generate realistic speech using a provided text prompt.

This skill generates realistic AI "talking head" videos from text prompts, useful for creating engaging content and demonstrations quickly.

Heygen generates professional-looking videos from text or images, streamlining content creation for marketing and communication needs.

Efficiently manages long conversations & complex tasks by intelligently prioritizing information within the context window.

This TTS skill converts text into natural-sounding speech, making your bot more engaging and accessible for users.

The Rag Engineer skill automatically constructs high-quality retrieval augmented generation (RAG) pipelines for seamless knowledge integration and improved AI responses.

Develop custom voice AI models, fine-tune existing ones, and optimize performance for unique applications.

Undocs removes unwanted documents from your chat history, streamlining conversations and protecting sensitive information.

Langfuse streamlines LLM workflows by providing a unified interface for prompt management and execution, boosting developer productivity.

Extracts text from images accurately using PaddleOCR, boosting OCR applications and enabling automated data extraction from visual sources.

Automates greptime setup using Ollama containers for local time series data analysis and visualization.

This skill provides insights into Transformer models’ decision-making by highlighting influential input tokens, aiding in debugging and understanding predictions.

Precisely segments images into distinct objects using a powerful, adaptable zero-shot segmentation model.

OpenRLHF Training enables fine-tuning of large language models using Reinforcement Learning from Human Feedback for improved alignment and helpfulness.

Shap generates creative prompts for various content types – ideal for sparking inspiration and overcoming writer's block.

Stable Baselines3 provides robust, easy-to-use implementations of reinforcement learning algorithms for training agents to master complex tasks.