Short lesson

Resources, Tools, and Prompts: The Three Core MCP Primitives

MCP exposes three distinct primitives: Resources (read-only file/database data), Tools (executable functions with side effects), and Prompts (reusable prompt templates).

Host ClientClaude / Cursor / IDE

Requests tool execution via JSON-RPC protocol.

TransportSTDIO / SSE Network

Transports encrypted payload with zero schema loss.

FastMCP ServerPython / TS Server

Validates Pydantic types & executes business logic.

ResourceEnterprise Database / API

Returns authenticated result to host context.

Technical Interview BattlecardInterview Battlecard

Interview Questions & Answers for Resources, Tools, and Prompts: The Three Core MCP Primitives

Curated model answers, architectural justification, and technical traps frequently tested in senior RPA technical rounds.

See Interview Q&A Based On This Topic

Go deeperResources, Tools, and Prompts: The Three Core MCP Primitives: Enterprise Masterclass

5 Parts

Chapters:5 Parts Available

PART 01Foundational Context

1 of 5

Core Architectural Foundation & Operational Principles

MCP exposes three distinct primitives: Resources (read-only file/database data), Tools (executable functions with side effects), and Prompts (reusable prompt templates). In enterprise systems design, Resources, Tools, and Prompts: The Three Core MCP Primitives represents a foundational capability within the Model Context Protocol (MCP) Enterprise Architecture domain. Unlike fragile deterministic scripts that crash when real-world inputs shift, autonomous systems achieve high resilience by coupling foundation reasoning models with strict tool contracts. Under the hood, this architecture balances token latency budgets, state persistence, and circuit-breaker guardrails to ensure mission-critical execution across production workloads.

Why this is critical

Clearly separating read-only context (Resources) from action execution (Tools) enforces security and least privilege.

Operational mechanics

Clients list resources via resources/list, read contents with resources/read, and execute functions via tools/call.

Practical application

Build an MCP server that exposes a text file as a Resource and a word-count function as a Tool.

Enterprise Production Standard

Enforces zero hardcoded credentials, explicit step ceilings, and automated telemetry logging.

PART 02Technical Breakdown

2 of 5

Visual Execution Architecture & Component Pipeline Canvas

This enterprise architecture canvas visualizes the exact state transitions, tool invocations, and guardrails required for Resources, Tools, and Prompts: The Three Core MCP Primitives:

n8n Canvas Workflow & Agent Execution Visualizer

Agent Flowchart

MODEL CONTEXT PROTOCOL (MCP) UNIVERSAL TOPOLOGY (Resources, Tools, and Prompts: The Three Core MCP Primitives):
┌────────────────────────────────────────────────────────────────────────┐
│ MCP Host Applications: Claude Desktop / Cursor / Antigravity IDE       │
└───────────────────────────────────┬────────────────────────────────────┘
                                    ▼ JSON-RPC 2.0 Protocol
┌────────────────────────────────────────────────────────────────────────┐
│ Transport Layer:                                                       │
│  ┌──────────────────────────────────┐  ┌─────────────────────────────┐ │
│  │ STDIO Transport (Local Machine)  │  │ SSE Transport (Remote Cloud)│ │
│  │ Zero network exposure / Unix pipe│  │ HTTPS / Server-Sent Events  │ │
│  └──────────────────────────────────┘  └─────────────────────────────┘ │
└───────────────────────────────────┬────────────────────────────────────┘
                                    ▼
┌────────────────────────────────────────────────────────────────────────┐
│ Enterprise FastMCP Servers:                                            │
│  ┌─────────────────────────┐  ┌──────────────────┐  ┌────────────────┐ │
│  │ Postgres / Snowflake DB │  │ GitHub / GitLab  │  │ Jira / SAP ERP │ │
│  │ Read/Write Tool Endpoints│  │ Codebase Tools   │  │ Ticketing API │ │
│  └─────────────────────────┘  └──────────────────┘  └────────────────┘ │
└────────────────────────────────────────────────────────────────────────┘

Swipe horizontally for full architecture⟷

PART 03Technical Breakdown

3 of 5

Fortune 500 Enterprise Case Study: Block (Square & Cash App)

At Block (Square & Cash App), handling 350,000 monthly internal engineer tool interactions, operational teams faced severe bottlenecks prior to implementing Resources, Tools, and Prompts: The Three Core MCP Primitives.

Operational Challenge

Engineering teams built separate plugins for Claude, ChatGPT, and IDE extensions, duplicating tool code across 12 repositories.

Architectural Solution

Consolidated all internal developer tools into enterprise FastMCP servers exposing database, deploy, and alerting primitives.

Quantifiable Business Outcome

Cut developer tooling maintenance by 75%, allowing engineers to securely query production databases directly from Cursor and Claude Desktop.

PART 04Technical Breakdown

4 of 5

Architectural Decision Matrix & Technical Comparison

Evaluate the trade-offs, SLAs, and performance metrics when deploying Resources, Tools, and Prompts: The Three Core MCP Primitives in production:

Evaluation Dimension	Proprietary API Plugins	Standard REST Endpoints	Model Context Protocol (MCP)
Interoperability	❌ Locked to one specific vendor	Requires custom client wrapper	✅ Universal 'USB-C' for all AI clients
Transport Flexibility	HTTP only	HTTP only	✅ STDIO (Local) + SSE (Remote Cloud)
Capability Discovery	Manual endpoint documentation	OpenAPI JSON specs	✅ Dynamic runtime protocol negotiation
Security & Sandboxing	Variable API token exposure	API token per service	✅ Scoped permissions & local process pipe
Ecosystem Adoption	Fragmented across ecosystems	Universal but manual	🟢 Standardized across Claude, Cursor, AGY

Scroll table sideways→

PART 05Key Takeaways

5 of 5

Senior AI Systems Architect Field Pitfalls & Production Standards

Avoid common architectural mistakes when operating autonomous agent workflows at enterprise scale:

Common Mistake to Avoid

Exposing data lookups as Tools with heavy side effects instead of lightweight read-only Resources.

Mandatory Production Rule

Use Resources for read-only static context and reserve Tools strictly for dynamic computations or external mutations.

Real-World Production Trap

Neglecting to set max_iterations on reasoning loops, allowing confused agents to enter infinite retry cycles and burn API budgets.

High-Volume Failure Mode

Omitting circuit breakers on external API tools, causing transient 5xx errors to permanently crash parent coordinator states.

Senior AI Agent Architect Field Insight

“Differentiate between Resources, Tools, and Prompts in the Model Context Protocol specification.”

Ready to test your technical mastery?Review model answers and candidate defense questions curated for Resources, Tools, and Prompts: The Three Core MCP Primitives.

Open RPA Fundamentals & Concepts Q&A

Agentic AI & n8n