AI Skills Directory
120 reusable AI skills and capabilities — prompt templates, agent tools, context frameworks, and composable modules for building autonomous workflows.
120 skills
Transfer Learning
by Community
Leverages knowledge from a source domain to improve model performance on a target domain with limited labeled data. A foundational technique for reducing training costs and accelerating model development across diverse applications.
Chain-of-Thought
by AaaS
Guides LLMs to produce step-by-step reasoning before arriving at a final answer. Dramatically improves performance on math, logic, and multi-step problems by making the model's reasoning process explicit and verifiable.
Prompt Engineering
by AaaS
The foundational discipline of crafting effective prompts to elicit desired behaviors from language models. Covers system prompt design, instruction formatting, output structuring, temperature tuning, and iterative prompt refinement techniques.
Code Generation
by AaaS
Generates functional code from natural language descriptions, specifications, or partial implementations. Covers multiple languages and frameworks with support for boilerplate scaffolding, algorithm implementation, and API integration patterns.
Function Calling
by AaaS
Enables LLMs to invoke external functions by generating structured JSON arguments matching defined schemas. Supports parallel function calls, error handling, and chained invocations for complex multi-step tool interactions.
Collaborative Filtering
by Community
Predicts user preferences by identifying patterns from collective user-item interaction histories, using memory-based neighborhood methods or model-based matrix factorization and neural approaches. The backbone of recommendation systems at scale across e-commerce, streaming, and social platforms.
Few-Shot Learning
by AaaS
Teaches LLMs to perform tasks by providing a small number of input-output examples in the prompt. Enables rapid task adaptation without fine-tuning by demonstrating the desired pattern through carefully selected, representative examples.
Tool Use
by AaaS
Equips AI agents with the ability to select and use appropriate tools from a defined toolkit to accomplish tasks. Covers tool selection logic, input marshalling, output interpretation, and fallback strategies when tools fail or return unexpected results.
Speech Recognition
by AaaS
Teaches integration and optimization of automatic speech recognition (ASR) systems — from Whisper to streaming cloud APIs — for agentic voice pipelines. Covers language identification, word error rate reduction, punctuation restoration, and handling noisy audio environments.
Time-Series Forecasting
by Community
Predicts future values of sequential, time-indexed data using classical statistical models (ARIMA, ETS), gradient boosting (LightGBM, XGBoost), and deep learning architectures (Transformers, N-BEATS, TFT). Handles trend, seasonality, exogenous covariates, and uncertainty quantification.
Domain-Specific Fine-Tuning
by Community
Adapts a general-purpose pretrained model to a narrow domain by continuing training on curated domain corpora or instruction datasets. Produces specialized models that outperform generalist baselines on domain-specific benchmarks while preserving broad language understanding.
Code Review
by AaaS
Analyzes code for bugs, security vulnerabilities, performance issues, and style violations. Provides actionable feedback with severity levels and suggested fixes aligned to language-specific best practices and project conventions.
Hybrid Recommendation Systems
by Community
Combines collaborative filtering and content-based signals — along with contextual, knowledge-graph, and session-based features — into unified ranking models that outperform single-strategy approaches. Modern implementations use two-tower neural architectures for efficient retrieval followed by cross-attention reranking.
Graph Neural Networks
by Community
Applies deep learning directly to graph-structured data by passing and aggregating messages between connected nodes across multiple layers, enabling node classification, link prediction, and graph-level tasks. Powers state-of-the-art knowledge graph completion, molecular property prediction, and social network analysis.
Reinforcement Learning for Control
by Community
Trains control policies for autonomous systems through environment interaction and reward signals using model-free (PPO, SAC, TD3) and model-based (MBPO, Dreamer) RL algorithms. Enables superhuman performance in complex continuous control tasks from locomotion to manipulation.
Summarization
by AaaS
Condenses long documents into concise summaries while preserving key information and maintaining factual accuracy. Supports extractive, abstractive, and hierarchical summarization with configurable length, style, and focus area parameters.
Anomaly Detection
by Community
Identifies unusual patterns, outliers, and change points in time-series and tabular data using statistical, density-based, isolation forest, autoencoder, and transformer-based methods. Fundamental for operational monitoring, fraud detection, and predictive maintenance systems.
RAG Retrieval
by AaaS
Retrieval-augmented generation skill enabling agents to query external knowledge bases via vector similarity search, reranking, and context injection for grounded, accurate responses.
Object Detection
by AaaS
Teaches agents to identify and localize multiple objects within images using bounding-box regression and classification heads. Covers model selection (YOLO, DETR, RT-DETR), confidence thresholding, NMS, and integrating detection pipelines into downstream agentic workflows.
Code Debugging
by AaaS
Diagnoses and resolves software bugs by analyzing error messages, stack traces, and code behavior. Applies systematic debugging strategies including root cause analysis, state inspection, and targeted fix generation with regression awareness.
Semantic Search
by AaaS
Enables meaning-based retrieval by converting queries and documents into dense vector representations and finding nearest neighbors. Foundational skill for any RAG pipeline or knowledge-base-powered agent.
Federated Learning
by Community
Trains machine learning models across decentralized data sources (devices or organizations) without centralizing raw data, using local computation and aggregated gradient updates. Enables collaborative model improvement while preserving data sovereignty and regulatory compliance.
Content-Based Recommendation
by Community
Recommends items by matching item feature profiles to user preference profiles derived from their interaction history, using TF-IDF, embeddings, and semantic similarity techniques. Effective for cold-start scenarios where user interaction data is sparse and item metadata is rich.
Text Classification
by AaaS
Categorizes text documents into predefined or dynamically generated classes using LLM-based inference. Supports multi-label classification, hierarchical taxonomies, and zero-shot classification without task-specific training data.
Build with composable AI skills
AaaS Academy packs bundle the best skills into ready-to-deploy action modules. Learn to build with AI agents in days, not months.
Explore Academy Packs