"DuckDB" matching MCP servers:

Enables read-only SQL querying and exploration of data files (CSV, Parquet, JSON, Excel, etc.) via DuckDB, supporting local paths, globs, URLs, and S3 buckets.

Enables querying and updating Markdown frontmatter metadata using DuckDB SQL, with optional semantic search capabilities for finding similar documents based on content.

Data-first MCP server for reading and writing spreadsheets (.xlsx, .csv, .ods) - cell-level ops + DuckDB SQL engine.

Query local CSV, Parquet, JSON and TSV files with real SQL via DuckDB. Gives your AI coding tool ground-truth data access instead of hallucinated answers.

MCP server for sandboxed, read-only SQL queries on CSV/Parquet/JSON files via DuckDB, limited to a specified directory.

Enables read-only access to explicitly approved local and remote filesystems, including inspecting Git repositories and querying SQLite and DuckDB databases, with root administration kept separate from MCP.

agentcairn is a local-first memory MCP server: your agent's memories live as Markdown in an Obsidian vault you own — the source of truth — with a rebuildable DuckDB index providing fast hybrid BM25 + vector + graph recall. It exposes tools to capture, recall, and manage those memories (non-lossy, with secret redaction) and works the same across Claude Code, Codex, Cursor, and any MCP host.