ParadeDB Skill

ParadeDB brings Elasticsearch-quality full-text search and analytics to Postgres using the pg_search extension. It provides BM25 ranking, advanced tokenizers, aggregations, and real-time indexing with ACID compliance.

Key capabilities: Full-text search with BM25 relevance scoring, hybrid search (keyword + semantic embeddings via pgvector), tokenization and analyzer configuration, fuzzy matching, phrase queries, faceted search, sorting, snippets and highlighting, and Tantivy-powered performance.

For complete and up-to-date ParadeDB documentation, always fetch:

https://docs.paradedb.com/llms-full.txt

This contains the full documentation optimized for LLMs. Use your web fetching tool to retrieve current docs before answering ParadeDB questions.

Network Failure Rules (Mandatory)

If llms-full.txt (or any required docs URL) cannot be fetched due to DNS/network/access errors:

1. State clearly that live docs could not be accessed and include the actual error.
2. Ask the user whether to proceed with local/repo-only context or to retry later.
3. Do not invent or infer doc URLs, page paths, or feature availability.
4. Do not present unverified links as real.
5. Label any fallback statements as assumptions and keep them minimal.

Never silently switch to guessed documentation structure when network access fails.