发现 devops llm ai 相关 Skills，比较来源仓库和用途，查看适合你工作流的技能说明。
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects the agent ('No, that's wrong...', 'Actually...'), (3) Use...
Core behavioral guidelines and interaction patterns for the agent. This skill provides the foundation for how the agent should communicate and behave across all interactions. Use this skill to establish consist...
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends the agent's capabilities with specialized knowledge, workflows, or...
Use for Rapidata Python SDK tasks: authenticate with RapidataClient, submit labeling orders from datasets (compare/classification/ranking/free-text/locate/draw/select-words/timestamp), apply filters/settings/se...
Ensures all AI-generated output fields have proper validation. Auto-activates on "AI 輸出", "LLM", "Gemini", "GPT", "truncate", "截斷" keywords. Lesson learned: 2026-01-08 Quick Feedback, Report, Deep Analyze t...
Monitors context token usage and warns when approaching limits. Automatically activates when context usage exceeds thresholds (30k, 50k, 70k tokens). Provides recommendations for context management and /clear u...
Add new instruction to the global CLAUDE.md file
Generate and edit images using Google Gemini's Nano Banana image generation API via undyapi.com proxy. Use when users want to generate images from text descriptions, edit existing images with AI, or create visu...
Turn experiment reports and notes into summaries and new/updated result skills.
Searches across Obsidian vault, synthesizes findings from multiple notes, and creates comprehensive research documentation. Gathers scattered information using filesystem search and dataview queries, analyzes c...
Probability-based text analysis providing GLTR token rank histograms, DetectGPT curvature probes, and Coh-Metrix-inspired cohesion metrics. Designed to compose with ai-check for comprehensive AI writing pattern...
Prepares meeting materials by gathering context from Obsidian vault, enriching with research, and creating comprehensive meeting documentation. Generates both internal prep notes with background and external ag...
Transforms conversations and discussions into structured documentation notes in Obsidian. Captures insights, decisions, and knowledge from chat context, formats with appropriate templates and frontmatter, and s...
Comprehensive SessionEnd hook that syncs and synthesizes all daily completions, notes, todos, and conversation data into unified daily notes and knowledge bases.
Information about Probitas framework. Use when asked "what is Probitas", explaining its purpose, features, or comparing with other test frameworks.
Prompt for creating custom agent files (.agent.md)
Prompt for creating custom prompt files
Group Relative Policy Optimization for reinforcement learning from human feedback. Covers GRPOTrainer, reward function design, policy optimization, and KL divergence constraints for stable RLHF training. Includ...
Fast inference with Unsloth and vLLM backend. Covers model loading, fast_generate(), thinking model output parsing, and memory management for efficient inference.
Supervised Fine-Tuning with SFTTrainer and Unsloth. Covers dataset preparation, chat template formatting, training configuration, and Unsloth optimizations for 2x faster instruction tuning. Includes thinking mo...
Open WebUI AI chat interface management via Podman Quadlet. Provides a web UI for interacting with Ollama models. Use when users need to configure, start, or manage the Open WebUI service.
Direct REST API operations for Ollama using the requests library. Covers all /api/* endpoints for model management, text generation, chat completion, embeddings, and streaming responses.
GPU monitoring and performance metrics for Ollama inference. Check GPU status, VRAM usage, loaded models, and inference performance metrics like tokens per second.
Transformer architecture fundamentals. Covers self-attention mechanism, multi-head attention, feed-forward networks, layer normalization, and residual connections. Essential concepts for understanding LLMs.
默认按已收录数据中的 GitHub 仓库 Stars 排序。Stars 属于来源仓库，同一仓库的多个技能可能共享相同数值；它反映仓库关注度，不代表单个技能的安装量或质量评分。