> For the complete documentation index, see [llms.txt](https://docs.siit.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.siit.io/integrations/knowledge-base.md).

# Knowledge Base

Connect Siit to your knowledge base so the documentation your teams already maintain — onboarding guides, how-tos, policies, troubleshooting steps

### Why connect a knowledge base

Your best knowledge already exists somewhere — in Notion, Confluence, or Slab. Connecting it to Siit means:

* **Instant AI answers.** The Triage Agent and the IT Agent search your connected knowledge to answer employee questions in Slack, Teams, and the Portal — with source references pointing back to the original page.
* **One-click inserts for agents.** Agents pull the right synced article into a reply from the composer's article picker, so the right resolution reaches the employee in seconds.
* **Self-service deflection.** Employees searching the Portal or asking the AI Agent get pulled into the right existing article before they even need to open a request.
* **Source of truth stays where it is.** Siit doesn't replace Notion, Confluence, or Slab — it reads from them, respects their permissions, and keeps content in sync as your teams edit the originals.
* **Zero content migration.** No CSV exports, no copy-pasting. Connect, pick the spaces you want synced, and Siit does the rest.

### Supported knowledge bases

Siit integrates natively with three knowledge platforms:

* **Notion** — sync selected workspaces, pages, or databases. Ideal for teams that already use Notion as their internal wiki. See the Help Center guide: [Notion integration](https://help.siit.io/notion-integration).
* **Confluence** — sync selected spaces from Atlassian Confluence Cloud. The go-to for teams standardized on the Atlassian suite. See the Help Center guide: [Confluence integration](https://help.siit.io/confluence-integration).
* **Slab** — sync selected topics and posts from Slab, for teams using it as their modern wiki.

Using a different knowledge base (GitBook, Guru, Document360, etc.)? Reach out via in-app chat — we're actively collecting interest to expand coverage.

### How the sync works

* **Initial import** — once you authorize and pick the spaces, collections, or topics you want synced, Siit pulls them in and converts each page into a Siit Article.
* **Continuous sync** — content refreshes on a schedule, so edits made in Notion, Confluence, or Slab flow into Siit automatically.
* **Source tracking** — every synced article displays its source (Notion, Confluence, or Slab) and links back to the original, so employees and agents can always jump to the canonical version.
* **Respects permissions** — Siit honors the access controls of your knowledge base. If a page is restricted, Siit respects that restriction in search and suggestions.

### Where connected knowledge shows up

Once synced, your knowledge powers every surface in Siit:

* **Portal Knowledge Base** — employees search and browse synced articles directly in the self-service portal.
* **Triage Agent** — suggests the relevant article to an employee before a request is created, with a summary and a link to the source.
* **IT Agent** — the AI Agent cites articles in Slack and Teams answers, with links back to the source.
* **Request replies** — agents insert articles directly into responses as rich cards.
* **Workflow actions** — attach articles to automated replies (e.g., "send password reset guide when reset request is submitted").

### What each integration covers

| Capability                             | Notion | Confluence | Slab |
| -------------------------------------- | ------ | ---------- | ---- |
| Sync pages / articles into Siit        | ✓      | ✓          | ✓    |
| Scope by space / workspace / topic     | ✓      | ✓          | ✓    |
| Automatic content refresh              | ✓      | ✓          | ✓    |
| Source link preserved on every article | ✓      | ✓          | ✓    |
| Used by AI agents for suggestions      | ✓      | ✓          | ✓    |
| Used in Portal Knowledge Base          | ✓      | ✓          | ✓    |
| Respects source permissions            | ✓      | ✓          | ✓    |

### Connecting a knowledge base

Each knowledge tool has its own connection flow, but the pattern is the same:

1. Go to **Settings → Workspace → Integration library** and find your knowledge tool in the library.
2. Authorize the connection — Notion and Slab use OAuth; Confluence uses an Atlassian API token plus your site URL.
3. Pick the scope — the specific workspaces, spaces, or topics you want Siit to read. Start small: one or two high-traffic spaces are plenty to begin with.
4. Run the initial sync and review the imported articles in **Context → Articles**.
5. Publish articles and make sure the Triage Agent and the IT Agent are enabled to use them.

Detailed step-by-step guides are available in the Help Center:

* [Notion integration setup](https://help.siit.io/notion-integration)
* [Confluence integration setup](https://help.siit.io/confluence-integration)

### Common use cases

* **AI-powered self-service.** Employees ask "How do I reset my Okta MFA?" in Slack. The IT Agent searches synced Notion pages, finds the matching guide, and answers with a source link — no ticket created.
* **Faster agent replies.** An agent opens a new request about VPN access and pulls the right Confluence page into the reply from the composer's article picker, in one click.
* **Portal knowledge browsing.** A new hire lands on the Siit portal and browses synced Notion onboarding guides, finding answers before their Day-1 questions pile up.
* **Deflection at submission.** When an employee starts submitting a "Password reset" service, Siit suggests a relevant synced article first — many resolve it themselves.
* **Content lifecycle alignment.** Your docs team edits a policy in Confluence. On the next sync, every Siit surface reflects the update — no manual re-publishing.

### Pair with AI for maximum impact

Knowledge integrations unlock the full value of Siit's AI features:

* **Triage Agent** — uses your connected knowledge to answer employees at intake and deflect the request when an article resolves it.
* **IT Agent** — cites synced articles as sources when answering employees in Slack, Teams, and the Portal.
* **Articles** — how synced content fits into the broader Siit knowledge model alongside articles written directly in Siit or saved from Slack.

The more high-quality content you sync, the better AI performs. Our recommendation: start with your 10–20 most-used internal docs, validate the AI suggestions against real questions, then expand the sync scope.

### Tips

* **Start with a focused scope.** Sync the one or two spaces where your highest-quality, most-used documentation lives. Quality beats quantity for AI accuracy.
* **Title articles as questions employees ask.** "How do I reset my Okta MFA?" retrieves better than "Okta MFA Reset Procedure."
* **Use dedicated service accounts for Confluence.** Atlassian API tokens are tied to a user account — a dedicated service user keeps the integration stable across team changes.
* **Review what AI is surfacing.** In **Context → Articles**, the analytics view shows which articles AI suggests most often. Low-quality or outdated articles can be demoted or retired.
* **Keep one source of truth per topic.** If the same content exists in Notion and Confluence, pick one, and retire duplicates. AI can't judge which version is canonical — you need to.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.siit.io/integrations/knowledge-base.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.