skills

Retrieve, analyse, and reason about personal health, training, well-being, and fitness data from Garmin Connect using the python-garminconnect library and the shared garmin_client package at ~/.garmin. Use this skill when the user asks to: - fetch or display health data (steps, heart rate, sleep, HRV, stress, body battery, SpO2, respiration, hydration) - analyse training or activity data (cycling, running, workouts, training load, training effect, HR zones) - query fitness metrics (VO2max, lactate threshold, intensity minutes, floors, calories) - write code that fetches data from Garmin Connect - debug Garmin Connect authentication (429 errors, rate limiting, token issues, browser cookies) - set up or install the garmin_client package

Send messages or media (images, audio, voice, video, animations, documents) to Telegram chats via bot API. Handles automatic message chunking for long text. Use when agent needs to send notifications, alerts, logs, status updates, text, or any media file to Telegram.

Transform PDFs into pristine markdown optimized for LLM consumption. Uses docling for initial conversion, then AI-driven iterative cleanup to produce publication-quality output with perfect structure, consistent formatting, and zero artifacts. Handles any PDF type through adaptive pattern recognition - no hardcoded rules. Fast processing with --no-ocr flag.

Semantic code search and codebase understanding through AI-powered indexing. Use for codebase exploration, pattern discovery, symbol lookup, impact analysis, DRY detection, and analysis of logs/git history. Triggered by needs for code understanding, searching similar implementations, finding patterns, or analyzing codebases semantically.