/ connectors · Airtable

Connect Airtable to Mari

Airtable bases quietly become systems of record: the vendor list, the launch checklist, the customer research log. Then someone asks a question the base could answer, and instead of querying it they ask the person who maintains it.

Mari's Airtable connector makes the base part of your team's searchable knowledge.

What syncs

• One document per table in the base you choose: the table name plus its records (up to 200 per table) rendered as readable field lists.
• Change detection by content. Airtable offers no change feed on this surface, so Mari says so instead of pretending. Every sync re-renders each table, and the pipeline's content hash skips whatever came out identical. Unchanged tables cost nothing to re-sync.

Connecting takes a minute

Two fields: an Airtable personal access token (pat...) and the base ID (app..., from the base's API docs). Test connection shows Airtable's actual error if something's off, verbatim. Connect, and a scheduled sync flow keeps the base fresh from then on.

What you get back

Ask a question the base can answer and the answer streams with numbered citations back to the exact table. Airtable records rank alongside your repos, Slack threads, and wikis in one hybrid search, so the spreadsheet-shaped knowledge stops being a silo.

Runs on your infrastructure

Mari is self-hosted and Apache 2.0. Postgres with pgvector is the whole data plane, models run locally through ollama, and every LLM feature has a deterministic fallback. Your base never leaves your deployment.

Start here

• Quick start: docker compose up to a working knowledge base.
• The docs cover connecting a source step by step.
• Book a demo to see it live.

Also connects: Notion, Trello, Google Drive, Slack, GitHub, and more. 12 connectors, one pipeline.