HKEx Filing Scraper — Docs

An open-source Python tool that scrapes 25+ years of Hong Kong Stock Exchange (HKEx) regulatory filings and ingests them into any of nine databases — with full-text and table extraction, chunk-level coverage, optional graph linking, and a hosted MCP gateway so AI agents can query HKEx live.

Live-MCP-Gateway{ .md-button .md-button--primary } Getting-Started{ .md-button } AI-Agent-Support{ .md-button } GitHub{ .md-button }

Start here

• Whats-New
• Getting-Started
• Sinks
• Configuration
• CLI
• MCP
• Live-MCP-Gateway — the hosted, no-install endpoint, with search filters and facet browsing
• AI-Agent-Support
• Try it locally (examples/)

Vendors & integrations

• Databases — nine first-class destinations; see the Sinks.
• AI clients — ready-made configuration for AI-Agent-Support.
• Available on — PyPI · Glama · MCP Registry.

Per-sink guides

Every sink is a first-class destination. Rows are in the documented popularity order.

• postgres — PostgreSQL
• mysql — MySQL-and-MariaDB
• sqlite — SQLite
• mongodb — MongoDB
• mariadb — MySQL-and-MariaDB
• neo4j — Neo4j
• clickhouse — ClickHouse
• duckdb — DuckDB
• surrealdb — SurrealDB

Operations

• Architecture
• Troubleshooting
• Testing
• De-risking
• Legal

Releasing

• Releasing
• Release-Automation
• Upgrading

Decisions

• ADR-0001-Versioning
• ADR-0002-Multi-Sink
• ADR-0003-Support-Policy
• ADR-0004-Publish-To-PyPI

Project

• Style
• README
• Contributing
• Changelog