aka. Agent Skills
Discover skills for AI coding agents. Works with Claude Code, OpenAI Codex, Gemini CLI, Cursor, and more.
Use this skill when working with Vision-Language Models. Covers visual encoder selection (CLIP/SigLIP/EVA), vision-language alignment, instruction tuning, multi-image conversation, and VLM evaluation.
View and manage hierarchical context for the current directory
提供通用的数据库查询工作流程、SQL 模板和结果分析方法。使用 exec_sql 工具执行查询。
Write docstrings for PyTorch functions and methods following PyTorch conventions. Use when writing or updating docstrings in PyTorch code.
Log and analyze workouts. Use when the user wants to (1) record a workout at the gym via conversation, (2) parse workout markdown files to CSV, (3) query exercise history with DuckDB, or (4) track fitness progress. Triggers on "log workout", "record workout", "track workout", "add exercises", "what did I lift", etc.
Manages temporary and defined agents including creation, promotion, cleanup, and namespacing. Use when user creates custom agents, asks about agent lifecycle, temp agents, or agent management.
Investment intelligence with power structure awareness. Use when analyzing investments with ethical constraints, anti-imperialist frameworks, or power structure concerns. Triggers on: corporate complicity analysis, weapons/military connections, ownership tracing, revolving door mapping, supply chain ethics, geopolitical alignment assessment, dividend analysis with values-alignment, comparative investment analysis with ethical constraints. Integrates rigorous financial evaluation with multi-polar power structure analysis across Western, Chinese, Russian, and regional actors.
Automates git commits, push, and PR creation with context-aware messages and ticket extraction. Use when performing git operations, creating commits/PRs, or when user mentions git, GitHub, commit, push, or pull request.
Master error handling patterns across languages including exceptions, Result types, error propagation, and graceful degradation to build resilient applications. Use when implementing error handling, designing APIs, or improving application reliability.
This skill helps Android developers integrate the Asleep SDK for sleep tracking functionality. Use this skill when developers ask about Android implementation, MVVM architecture patterns, permission handling, Jetpack Compose UI, Kotlin coroutines integration, or Android-specific sleep tracking features. This skill provides working code examples from the official sample app.
Use this skill when working with Hugging Face Transformers library. Covers model loading, fine-tuning, LoRA/QLoRA adaptation, tokenizer usage, datasets processing, and Trainer API.
Data Tool Suite - LLM-observable workflows for dbt project data exploration
Generate Mission Control ScrapeConfig YAML from natural language. Use when users ask to build a scraper, create a config scraper, or want YAML to scrape Kubernetes/AWS/GCP/Azure/SQL/HTTP/File/Exec/Logs/Slack/GitHub Actions/Trivy/Terraform sources.
Launch ralph.sh in background. Automatic execution until completion or critical blocker.
View and modify ConKeeper memory configuration settings. Use to adjust token budget, output style, and other preferences after memory initialization.
"Is this ACTUALLY verified, or did I just say it is?" - Ralph-wiggum methodology applied to factual accuracy. Use when (1) claims require evidence not assumption, (2) verification must be demonstrable, (3) single-pass investigation insufficient, (4) factual accuracy is critical. Provides the loop logic: iterate until verification thresholds met. Does NOT trigger for: opinions, preferences, how-to instructions, or when user explicitly wants quick answer.
Design and implement event stores for event-sourced systems. Use when building event sourcing infrastructure, choosing event store technologies, or implementing event persistence patterns.
Architecture phase (ARCHITECT) - transforms research into rigorous technical architecture through 5 mandatory design artifacts. Interactive and iterative.
Use this skill when implementing unified understanding and generation models. Covers multi-task architectures, autoregressive vs diffusion approaches, multimodal tokenization, and next-token prediction paradigms.
提供通用的SQL查询方法和数据库查询基础。使用 exec_sql 工具执行查询。