/ connectors · Slack

Connect Slack to Mari

The best answer your team ever wrote is sitting in a Slack thread nobody can find. Someone asked, someone who knew typed the real answer, everyone said thanks, and the correction sank below the scroll line by lunch. Slack holds an enormous share of your team's real knowledge, and almost none of it is retrievable when the question comes back.

Mari's Slack connector turns that history into a knowledge base your team can actually query.

What syncs

• Public-channel history, imported with a bot token. Messages group into one searchable document per channel per day, with timestamps and authors preserved, so a hit lands you in the right conversation, not a single orphaned message.
• Incremental sync. Re-syncs pick up from the newest message seen, and a day's document is always rebuilt complete, so nothing is half-captured.
• Nothing re-embedded that didn't change. Every connector feeds Mari's one ingestion pipeline: chunk, content-hash, embed. A no-op resync costs seconds.

Connecting takes a minute

Pick Slack from the connector list, paste a bot token, and hit Test connection. A wrong token shows Slack's actual error, verbatim, not "something went wrong." Connect, watch the first sync run live, and Mari quietly creates a scheduled sync flow for the source: visible and editable, not a hidden cron.

What you get back

Ask a question and the answer streams with numbered citations back to the channel and day it came from. Slack messages rank alongside your repos, wikis, and tickets in one hybrid search (keyword plus semantic, scored together). And the claims worth keeping can graduate. A thread's answer becomes a fact with an owner, a verification state, and a freshness date, so the next person gets the vetted version instead of the archaeology.

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 Slack history 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: Confluence, Notion, Jira, GitHub, Google Drive, Linear, Zendesk, and more. 12 connectors, one pipeline.