Connect Jira, Confluence, Notion or Linear and Khint keeps a local, always-fresh index of your work. Agents answer with your real tickets and pages. Never on our servers.
Switch a source on in Memory · Knowledge. Khint syncs it incrementally and the Test bar shows the exact excerpts an agent would read.
Type anything into Test the index and see what comes back, matched terms highlighted. What you see is byte-for-byte what gets injected into a run.
Same ask, two answers. Without the index: filler. With it: your ticket keys, your decisions, your numbers.
The success pill says “knowledge context used” when the index contributed — the loop is visible, never silent.
Enterprise AI platforms sync a copy of your corpus to their cloud and search it there. Khint inverts the model: the index lives with you.
JiraIssues: key, summary, status, description. Incremental on `updated`.
ConfluencePages: title and body, per space, most recently modified first.
NotionPages: title and block content, ordered by last edit.
LinearIssues: identifier, title, state, description.
The index is SQLite on your Mac. No embeddings service, no sync infrastructure, no copy of your corpus on anyone's cloud — including ours.
Each user indexes with their own account. What their Jira can't see, their index doesn't hold. No central access-control re-mapping to get wrong.
Sync makes no AI calls. Retrieval is local full-text search in milliseconds. Only the few matched excerpts ride along with runs that opt in.
On your Mac, in a local SQLite full-text index — never on Khint's servers. Unlike cloud platforms that keep a synced copy of your Confluence or CRM on their infrastructure, Khint holds no copy of your corpus anywhere. Switching a source off, or forgetting an integration's credentials, deletes its local index immediately.
Only the top 5 excerpts that match the text of the run you launched — a few hundred words, sent inside a tagged data block, the same way your active session context travels today. The index itself never leaves your machine, and retrieval is pure local search with no AI call.
Jira, Confluence, Notion and Linear at launch — each toggles independently from Memory's Knowledge section. Sync is incremental: after the first pass, only documents that changed since the last sync are fetched, every 30 minutes or on demand with Sync now.
No. Indexing makes no AI calls, and retrieval is local BM25 full-text search that answers in milliseconds. The only cost is the few hundred extra words added to the agent runs that opt in — usually absorbed by the normal per-run rounding.
Khint indexes with each user's own credentials, so every person's index contains exactly what their Jira or Confluence account can see — nothing more. There is no central copy and no access-control re-mapping: the tools' own permissions are the permissions.
Guided walkthroughs that run on this pillar, start to finish.
