Skip to main content

Sync anything to Open WebUI Knowledge Bases

Project description

📚 oikb

Keep your Open WebUI Knowledge Bases in sync. Point it at a local directory, a GitHub repo, a Confluence space, an S3 bucket, or any of 44 supported sources. Only new and modified files are uploaded via incremental SHA-256 diffing.

[!IMPORTANT] Requires Open WebUI 0.9.6+

Quick Start

pip install oikb

export OPEN_WEBUI_URL=http://localhost:3000
export OPEN_WEBUI_API_KEY=sk-your-api-key

# Sync a directory to a Knowledge Base
oikb sync ./docs --kb-id your-kb-id

# Or watch for changes and auto-sync continuously
oikb watch ./docs --kb-id your-kb-id

Commands

Command Description
oikb sync <source> Incremental sync to a Knowledge Base
oikb watch <dir> Watch for changes and auto-sync
oikb daemon Long-lived scheduler with HTTP API
oikb diff <source> Preview what a sync would do
oikb history View sync history
oikb ls List files in a Knowledge Base
oikb status Show KB info and file count
oikb reset Delete all files in a Knowledge Base
oikb config Manage saved URL and API key

Daemon

Run oikb daemon for production deployments. Reads .oikb.yaml and syncs each source on a schedule.

oikb daemon --port 8080

Features:

  • Scheduled sync — configurable per-source intervals (30m, 1h, 6h)
  • Webhooks — instant sync on push via /webhooks/github, /webhooks/gitlab, /webhooks/slack, /webhooks/confluence
  • Health checksGET /health for Docker/K8s readiness probes
  • Sync historyGET /history queryable log of all syncs
  • On-demand syncPOST /sync/{source} to trigger immediately
  • OpenAPI tool server — add http://oikb:8080 as a Tool Server in Open WebUI (Settings → Connections) and let the LLM trigger syncs, check status, and query history
# .oikb.yaml
sources:
  - source: github:owner/repo
    kb-id: team-wiki
    interval: 1h
    webhook: true

  - source: confluence:ENG
    kb-id: handbook
    interval: 6h

Docker Compose

services:
  open-webui:
    image: ghcr.io/open-webui/open-webui:main
    ports:
      - "3000:8080"

  oikb:
    image: ghcr.io/open-webui/oikb:latest
    environment:
      - OPEN_WEBUI_URL=http://open-webui:8080
      - OPEN_WEBUI_API_KEY=${OPEN_WEBUI_API_KEY}
    volumes:
      - ./.oikb.yaml:/app/.oikb.yaml:ro
    command: daemon
    ports:
      - "8080:8080"
    depends_on:
      - open-webui
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8080/health/ready"]
      interval: 30s
      timeout: 5s

44 Connectors

Category Sources
Code Repos GitHub, GitLab, Bitbucket
Cloud Storage S3, GCS, Azure Blob, Dropbox, R2, Google Drive, SharePoint, Egnyte, Oracle Cloud
Wikis & KBs Confluence, Notion, BookStack, Discourse, GitBook, Guru, Outline, Slab, Document360, DokuWiki, Google Sites
Ticketing Jira, Linear, Zendesk, Freshdesk, Asana, ClickUp, Airtable, ServiceNow, ProductBoard
Messaging Slack, Discord, Microsoft Teams, Gmail, Zulip
Meetings Gong, Fireflies
Forums XenForo
Sales & CRM Salesforce, HubSpot
Web Website / Sitemap crawler
oikb sync github:owner/repo --kb-id your-kb-id
oikb sync confluence:ENG --kb-id your-kb-id
oikb sync s3://bucket/prefix --kb-id your-kb-id
oikb sync servicenow:incident --kb-id your-kb-id

Some connectors need an optional extra: pip install oikb[gdrive], pip install oikb[s3], or pip install oikb[all] for everything.

Multi-KB Routing

Route files from a single source to different Knowledge Bases by glob pattern:

sources:
  - source: github:owner/repo
    routes:
      "docs/**/*.md": docs-kb
      "src/**": code-kb

Selective Sync Filters

Narrow what gets synced with include/exclude globs:

sources:
  - source: github:owner/repo
    kb-id: docs-only
    filter:
      include: ["docs/**/*.md", "*.txt"]
      exclude: ["drafts/**"]

Configuration

Resolved in order (highest priority wins):

  1. CLI flags (--url, --token)
  2. Environment variables (OPEN_WEBUI_URL, OPEN_WEBUI_API_KEY)
  3. Config file (~/.config/oikb/config.yaml)

History

oikb history                    # Table view
oikb history --json             # JSON output
oikb history --errors           # Failed syncs only
oikb history --clear --days 7   # Prune old entries

GitHub Actions

- name: Sync docs to Open WebUI
  uses: docker://ghcr.io/open-webui/oikb:latest
  with:
    args: sync /github/workspace/docs --kb-id ${{ secrets.KB_ID }}
  env:
    OPEN_WEBUI_URL: ${{ secrets.OPEN_WEBUI_URL }}
    OPEN_WEBUI_API_KEY: ${{ secrets.OPEN_WEBUI_API_KEY }}

How It Works

  1. Scan source, compute checksums
  2. Send manifest to Open WebUI /sync/diff
  3. Delete stale files, create missing directories
  4. Upload only new and modified files

License

MIT. See LICENSE for details.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

oikb-0.2.1.tar.gz (50.8 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

oikb-0.2.1-py3-none-any.whl (89.9 kB view details)

Uploaded Python 3

File details

Details for the file oikb-0.2.1.tar.gz.

File metadata

  • Download URL: oikb-0.2.1.tar.gz
  • Upload date:
  • Size: 50.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.8.0

File hashes

Hashes for oikb-0.2.1.tar.gz
Algorithm Hash digest
SHA256 19036fa8fd7531cc6f9d01996710a80374df19ab978a615180d5442a37fd0aef
MD5 f99f5ab07d243c17cfd932e89503603f
BLAKE2b-256 30cc20ff950c1577a7e5cf2b926600d1ea9507637938c7e25f0996ca04e9c325

See more details on using hashes here.

File details

Details for the file oikb-0.2.1-py3-none-any.whl.

File metadata

  • Download URL: oikb-0.2.1-py3-none-any.whl
  • Upload date:
  • Size: 89.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.8.0

File hashes

Hashes for oikb-0.2.1-py3-none-any.whl
Algorithm Hash digest
SHA256 5f65c0e8107564686e30148874a02af9a03e64cc3eaa3d7b080ff1c4c96a91cd
MD5 b346cbeea2d94eea2f10de74472f34ce
BLAKE2b-256 54666d1e01eb0ced905a93b14a8b0baa9d3e52a1fb0aa1f4fde0cce712bd430e

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page