Getting Started
Install metawarc and index, query, or replay a WARC collection.
Bounded indexing, extraction, and analysis for WARC collections
Index WARC collections into a versioned DuckDB catalog with Parquet sidecars, then query, extract, analyze, and replay without copying payloads.
pip install metawarc metawarc index 'archives/**/*.warc*' --dbfile collection.db --resume
Start with a section below, or use the sidebar from any page. The CLI entry point is metawarc. Source WARC files stay immutable; the DuckDB catalog and Parquet sidecars are the operational index.
Install metawarc and index, query, or replay a WARC collection.
End-to-end examples for indexing, export, analysis, replay, and agents.
Command-by-command reference for index, query, extract, analyze, and serve.
Workspace schema, typed queries, and the threat model for hostile archives.
Read-only REST API, local website replay, and MCP tools for agents.
Contributing, release checklist, community, and license.