# Ohu Web Intelligence & Data Extraction Engine

> High-performance Web Intelligence, Synthetic Dataset Generator, Vector RAG Chunker, Prompt Injection Security Auditor, and Answer Engine Optimization (AEO) Engine.

Ohu provides developer APIs and a native Remote Model Context Protocol (MCP) server to convert web pages, PDFs, and DOM structures into clean LLM context, fine-tuning datasets, and security audit reports.

## Main Pages & Developer Console
- [Home Page](https://sluxia.com/ohu): Interactive multi-tool console, live sandbox, pricing tiers, and platform features.
- [Developer Documentation](https://sluxia.com/ohu/docs): Complete API reference, cURL, Node.js, and Python code examples for all 18 production REST endpoints.
- [Remote MCP Server Console](https://sluxia.com/ohu/mcp-server): Model Context Protocol (MCP) Streamable HTTP configuration guide for OpenCode CLI, Claude Desktop, Cursor, Windsurf, Continue.dev, and Zed.
- [Developer Console Dashboard](https://sluxia.com/ohu/dashboard): API key management, real-time request logs, quota tracking, and subscription billing.

## Core API Endpoints & Capabilities
- [Synthetic Dataset Generator](https://sluxia.com/ohu/docs#endpoint-synthetic-dataset): Generate factual Q&A instruction pairs in JSONL format for OpenAI ChatML, Alpaca, ShareGPT, and DPO fine-tuning.
- [Indirect Prompt Injection Scanner](https://sluxia.com/ohu/docs#endpoint-security-audit): Audit web pages for hidden CSS injection attacks (`display:none`, `opacity:0`), zero-font text, and leaked API keys with NFKC normalization.
- [GEO Competitor Audit](https://sluxia.com/ohu/docs#endpoint-geo-competitor): Perform head-to-head AI recommendation audits comparing domain semantic density and citation probability against competitors.
- [Answer Engine Optimization (AEO)](https://sluxia.com/ohu/docs#endpoint-aeo): Analyze AI crawler access (GPTBot, ClaudeBot, PerplexityBot) and generate instant Organization / FAQ JSON-LD schemas.
- [Vector RAG Chunker](https://sluxia.com/ohu/docs#endpoint-rag-chunk): Chunk HTML/text into vector-ready blocks with preserved header ancestry (`H1 > H2 > H3`) and export formats for Pinecone, Qdrant, Weaviate, and Chroma.
- [HTML Table Parser](https://sluxia.com/ohu/docs#endpoint-table-extract): Parse HTML `<table>` elements into structured JSON row objects, GFM Markdown tables, and CSV format.
- [Article Body Distiller](https://sluxia.com/ohu/docs#endpoint-clean-article): Reader-mode article distiller that prunes navigation boilerplate and returns clean body text, lead summary, and hero image.
- [Live Web Search](https://sluxia.com/ohu/docs#endpoint-search): Multi-provider web search engine (Bing, DDG, Wikipedia) with domain authority classification and snippet proximity highlighting.
- [PDF to Markdown](https://sluxia.com/ohu/docs#endpoint-pdf-to-markdown): Download and convert PDF documents into structured Markdown text.

## Legal & Compliance
- [Privacy Policy](https://sluxia.com/ohu/privacy): EU GDPR and US CCPA/CPRA privacy disclosures, data processing roles, and in-memory transient data retention policy.
- [Terms of Service](https://sluxia.com/ohu/terms): Acceptable use policy, API quota limits, billing cycles, SLA liability limits, and dispute resolution.
- [Security Policy](https://sluxia.com/ohu/security): SOC2-aligned security controls, TLS 1.3 transport encryption, SHA-256 API key hashing, and responsible vulnerability disclosure policy.

## Full Context Manifest
- [Full LLM Manifest](https://sluxia.com/ohu/llms-full.txt): Comprehensive API schemas, JSON request payloads, and remote MCP tool definitions.