FrootAI — AmpliFAI your AI Ecosystem Get Started

FAI Glossary

FAI Protocol terms, FROOT taxonomy, WAF pillars, and 200+ AI/ML terms — the complete reference.

L12·10 min read·Reference

FROOT Taxonomy

FROOT is the 5-layer knowledge architecture that organizes all AI concepts from fundamentals to production. Each letter maps to a learning layer, and each layer contains 2-6 modules that cover the full AI lifecycle:

F— Foundations

Tokens, embeddings, model architectures, GPU compute. Modules F1 (GenAI Foundations), F2 (LLMs & SLMs), F3 (Glossary), F4 (.github Agentic OS). The bedrock layer — everything else builds on this.

R— Reasoning

Prompt engineering, RAG pipelines, deterministic AI patterns. Modules R1 (Prompt Engineering), R2 (RAG), R3 (Deterministic AI). The intelligence layer — how models think and retrieve knowledge.

O— Orchestration

Semantic Kernel, multi-agent systems, MCP tools, Azure AI services. Modules O1 (Semantic Kernel), O2 (AI Agents), O3 (MCP & Tools), O4 (Azure AI Foundry), O5 (Infrastructure), O6 (GitHub Copilot). The wiring layer — connecting models to the real world.

O— Operations

Azure AI Foundry, GPU infrastructure, Copilot integration. Shared with the Orchestration layer. The delivery layer — making AI solutions run at scale in production.

T— Transformation

Fine-tuning, responsible AI, production readiness. Modules T1 (Fine-Tuning), T2 (Responsible AI), T3 (Production). The maturation layer — from prototype to enterprise-grade.

FAI Protocol Terms

The FAI terminology stack — from specification to distribution:

TermDefinition
FAI ProtocolThe context-wiring specification. Defined in fai-manifest.json. Connects agents, skills, instructions, hooks, plugins, and guardrails into a unified play.
FAI LayerThe conceptual glue between primitives. MCP handles tool calling, A2A handles delegation, AG-UI handles rendering — the FAI Layer handles wiring.
FAI EngineThe runtime that reads fai-manifest.json, resolves dependencies, wires primitives, runs evaluations, and reports status.
FAI FactoryThe CI/CD pipeline — 20+ scripts and 15+ workflows that validate, generate, sync, and publish the ecosystem.
FAI PackagesDistribution channels: npm (frootai), PyPI (frootai-sdk), Docker, VS Code extension, GitHub Action.
FAI ToolkitThe 3 sub-kits: DevKit (agents, instructions, skills), TuneKit (openai.json, guardrails.json), SpecKit (schemas, validation).
FAI MarketplaceThe discovery registry — 77+ plugins, searchable, installable via CLI or VS Code. Plugins are LEGO blocks that snap into any play.

Play & Primitive Terminology

TermDefinition
Solution PlayA complete AI solution template — manifest + primitives + infrastructure + config. Numbered 01-68+.
PrimitiveA building block: agent, instruction, skill, hook, or plugin. Primitives are standalone but auto-wire when used inside plays.
Agent (.agent.md)A scoped persona with YAML frontmatter — description, model, tools, WAF alignment. Invoked via @mention in Copilot Chat.
Instruction (.instructions.md)Auto-attached context with applyTo glob patterns. Silently enriches Copilot for matching file types.
Skill (SKILL.md)Multi-step workflow with tool access. Folder-based: skill-name/SKILL.md. Invoked via #skill reference.
Hook (hooks.json)Event-driven automation. Fires on sessionStart, sessionEnd, userPromptSubmitted, preToolUse. Runs scripts automatically.
Plugin (plugin.json)Packaged bundle of primitives. Installable from marketplace. Contains manifest, README, and all wired assets.

Well-Architected Framework (WAF) Pillars

Every FAI solution play aligns to these 6 architectural pillars:

Security

Managed Identity, Key Vault, RBAC, content safety filters, network isolation, secret scanning. Zero hardcoded credentials.

Reliability

Retry with exponential backoff, circuit breakers, health checks, graceful degradation, multi-region failover.

Cost Optimization

Model routing (cheap model first, escalate if needed), token budgets, semantic caching, right-sizing compute.

Operational Excellence

CI/CD pipelines, IaC (Bicep/Terraform), structured logging, alerting, incident management, runbook automation.

Performance Efficiency

Streaming responses, async processing, caching layers, batch operations, CDN for static assets, connection pooling.

Responsible AI

Content safety, groundedness evaluation, fairness testing, transparency logging, human-in-the-loop, PII detection.

AI/ML Terms — Models & Architecture

TermDefinition
LLMLarge Language Model — neural network with billions of parameters trained to predict text. GPT-4o, Claude, Llama.
SLMSmall Language Model — compact model (1-7B params) optimized for specific tasks. Phi-3, Mistral 7B.
TransformerThe attention-based architecture behind modern LLMs. Self-attention enables parallel processing of token sequences.
TokenThe atomic unit of text for LLMs. ~4 characters in English. Both input and output are measured in tokens.
Context WindowMaximum tokens a model can process in a single request. GPT-4o: 128K tokens. Larger windows enable more context.
EmbeddingDense vector representation of text in high-dimensional space. Enables semantic similarity search. Key to RAG.
Fine-TuningTraining an existing model on domain-specific data to improve task performance. LoRA, QLoRA, full fine-tuning.
LoRALow-Rank Adaptation — efficient fine-tuning that trains small adapter matrices instead of full model weights.

AI/ML Terms — Inference & Generation

TermDefinition
TemperatureControls output randomness (0.0–2.0). Lower = deterministic, higher = creative. Use 0.0 for classification, 0.7 for chat.
Top-PNucleus sampling — selects from the smallest set of tokens whose cumulative probability exceeds P. Alternative to temperature.
Top-KLimits sampling to the K most probable tokens. Reduces hallucination risk at the cost of diversity.
HallucinationWhen a model generates plausible but factually incorrect content. Mitigated by RAG, grounding, and evaluation.
GroundingAnchoring model outputs to verifiable source data. Groundedness score measures factual alignment.
Prompt InjectionAttack where malicious input tricks the model into ignoring instructions. Mitigated by content safety filters and input validation.
StreamingSending tokens to the client as they are generated instead of waiting for full completion. Improves perceived latency.
Semantic CachingCaching LLM responses keyed by embedding similarity rather than exact string match. Reduces cost and latency for similar queries.

AI/ML Terms — RAG & Retrieval

TermDefinition
RAGRetrieval-Augmented Generation — fetches relevant documents before generating, grounding responses in real data.
ChunkingSplitting documents into smaller segments for embedding indexing. Strategies: fixed-size, semantic, sliding window.
Vector StoreDatabase optimized for similarity search over embeddings. Azure AI Search, Cosmos DB vector, Pinecone, Weaviate.
Hybrid SearchCombining keyword (BM25) and vector similarity search for better recall. Azure AI Search supports this natively.
RerankingSecond-pass scoring of retrieved documents using a cross-encoder model. Improves precision after initial retrieval.
MCPModel Context Protocol — open standard for connecting AI models to external tools and data sources. Replaces custom tool integrations.
A2AAgent-to-Agent protocol — enables delegation between agents. Complements MCP (tools) and AG-UI (rendering).

AI/ML Terms — Agents & Orchestration

TermDefinition
Semantic KernelMicrosoft's open-source AI orchestration SDK. Manages plugins, planners, and memory for agent applications.
Function CallingLLM capability to select and invoke structured functions. The model outputs a JSON tool call instead of text.
ReActReasoning + Acting pattern — the agent reasons about a step, takes an action, observes the result, then reasons again.
Chain-of-ThoughtPrompting technique where the model shows its reasoning step by step. Improves accuracy on complex tasks.
Multi-AgentArchitecture where multiple specialized agents collaborate. Builder/reviewer/tuner is a 3-agent chain.
GuardrailsRuntime safety checks on model inputs and outputs. Content safety, PII detection, prompt injection resistance.
AG-UIAgent-Generated UI protocol — standard for agents to render structured UI components (tables, charts, forms).

Evaluation & Quality Metrics

MetricScaleDefinition
Groundedness1–5Are claims supported by the retrieved context? The most critical RAG metric.
Relevance1–5Does the response address the user's question? Measures answer alignment.
Coherence1–5Is the response well-structured and logically organized?
Fluency1–5Is the language natural and grammatically correct?
Safetypass/failDoes the response avoid harmful, biased, or inappropriate content?