Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/TechFernandesLTDA/apex-mcp/llms.txt

Use this file to discover all available pages before exploring further.

apex-mcp is a Model Context Protocol (MCP) server that exposes 116 tools for building Oracle APEX 24.2 applications through AI assistants. Instead of navigating the APEX App Builder UI — clicking through wizards to create pages, regions, and items one at a time — you describe what you want in plain language and the AI does the work. Built on FastMCP 3.x and Python 3.11+, apex-mcp connects to Oracle Autonomous Database via the oracledb thin driver with mTLS wallet authentication. No Oracle Instant Client required. It works with Claude Code, Claude Desktop, GPT via OpenAI Agents SDK, Google Gemini CLI, Cursor, VS Code, and any other MCP-compatible AI client.

​
How It Works

apex-mcp creates APEX applications by generating and executing wwv_flow_imp_page.* PL/SQL calls directly against your Oracle database — the same internal mechanism APEX itself uses for import/export. This bypasses the App Builder UI entirely and is fully compatible with APEX 24.2.13 / Universal Theme 42. Your AI client sends a natural language prompt, apex-mcp translates it into the appropriate sequence of tool calls, and each tool generates and executes the corresponding PL/SQL. 
┌─────────────────────────────────────────────────────────────────┐
│  AI Client (Claude / GPT / Gemini / Cursor / VS Code)           │
│                                                                  │
│  "Create a full HR app from the EMPLOYEES table"                 │
└────────────────────────────┬────────────────────────────────────┘
                             │ MCP (stdio or HTTP)
                             ▼
┌─────────────────────────────────────────────────────────────────┐
│  apex-mcp server (FastMCP 3.x, Python 3.11+)                    │
│  116 tools across 17 categories                                  │
└────────────────────────────┬────────────────────────────────────┘
                             │ oracledb (mTLS)
                             ▼
┌─────────────────────────────────────────────────────────────────┐
│  Oracle Autonomous Database 23ai                                 │
│  wwv_flow_imp_page → APEX 24.2 App Builder                      │
└─────────────────────────────────────────────────────────────────┘
Every app build follows a consistent lifecycle: apex_connect() → apex_create_app() → add pages, regions, items, and components → apex_finalize_app(). An ImportSession singleton tracks all cross-references (region IDs, item names, page sequences) so tools can resolve dependencies automatically. Tools respect dry-run mode (log PL/SQL without executing) and batch mode (queue multiple statements for a single database round-trip).

​
Key Capabilities

116 Tools, 17 Categories

Full coverage from connection management and schema introspection through page creation, component assembly, chart generation, JavaScript injection, DevOps exports, and user management. Every APEX concept is a first-class tool.

High-Level Generators

apex_generate_from_schema builds a complete app from a list of tables — CRUD pages, dashboard, and navigation — in one call. apex_generate_crud, apex_generate_dashboard, apex_generate_wizard, and apex_generate_modal_form handle common patterns without manual assembly.

20 UI Components + 10 Chart Types

Rich HTML components (hero banners, KPI rows, progress trackers, leaderboards, heatmap grids, activity streams) rendered via PL/SQL region source. Ten advanced chart types beyond basic JET charts: stacked, combo, Pareto, scatter, range, animated counters, gradient donut, and bubble.

Three Transport Modes

stdio for local clients (Claude Code, Cursor, VS Code, Gemini CLI). streamable-http for remote or multi-client setups (OpenAI Agents SDK, Codespaces). SSE for clients that require Server-Sent Events. All controlled via CLI flags or environment variables.

Schema Introspection with FK Detection

apex_list_tables, apex_describe_table (with per-session caching), and apex_detect_relationships give the AI a complete picture of your database schema — including foreign key relationships — before generating any code.

Batch Mode + Dry-Run Mode

apex_begin_batch() / apex_commit_batch() queue multiple PL/SQL statements for a single database round-trip, dramatically reducing latency for large builds. apex_dry_run_preview(enabled=True) logs all generated PL/SQL without executing — useful for review before committing changes.

​
Requirements

ComponentVersion
Python3.11+
FastMCP3.0.0+
oracledb2.0.0+
Oracle APEX24.2 (Universal Theme 42)
Oracle DB19c+ (tested on ADB 23ai)
Oracle WalletRequired for mTLS connections
No Oracle Instant Client is required. The oracledb thin driver handles mTLS wallet connections natively in pure Python.

​
Next Steps

Quickstart

Install apex-mcp, configure your Oracle credentials, connect to Claude Code, and build your first APEX app in five minutes.

Configuration

Full reference for all Oracle connection variables, APEX workspace settings, and optional transport configuration.

Build docs developers (and LLMs) love

Get started for freeTalk to us