GitHub logo

Train your AI on GitHub

Auth: Personal Access TokenIncremental syncFull refresh

Connect your GitHub repositories to train your AI agent on your technical documentation. Open syncs README files, wikis, and markdown documentation to give your AI agent deep understanding of your codebase and processes.

Connect once, and Open automatically keeps your AI agent's knowledge up to date. When you update content in GitHub, the changes sync automatically—no manual retraining required.

GitHub logo→syncs to→AI Agent

What can be synced

Repositories — Repository metadata including descriptions, topics, and basic information.

Commits — Commit history with messages, authors, and timestamps from specified branches.

Issues — GitHub issues with titles, bodies, labels, assignees, and comments.

Pull Requests — Pull requests with descriptions, review status, commits, and review comments.

Comments — Issue comments and commit comments for discussion context.

Branches — Branch information and protection rules from repositories.

Features

  • •Repository Docs — Sync README files and documentation from selected repositories.
  • •Wiki Integration — Import GitHub Wiki pages for comprehensive documentation coverage.
  • •Markdown Support — Full markdown parsing with code block and formatting preservation.
  • •Branch Selection — Choose which branches to sync documentation from.

Requirements

  • •Personal Access Token or OAuth credentials
  • •List of repositories to sync (format: org/repo)
  • •Appropriate permissions for private repositories

How to connect

  1. 1.Go to GitHub Settings → Developer settings → Personal access tokens
  2. 2.Generate a new token with repo scope (and org:read for organization repos)
  3. 3.Copy the token for use in Open
  4. 4.In Open, go to AI Training → Sources → Connect GitHub
  5. 5.Enter your token and specify repositories (e.g., "org/repo" or "org/*" for all)
  6. 6.Optionally specify branches to sync (default branch is used if not specified)

Good to know

  • Use multiple tokens separated by "," to load balance API quota
  • Most streams support incremental sync (append + deduplication)
  • Specify branches like "org/repo/branch1,branch2" to sync specific branches
  • Invalid or inaccessible repository names are skipped with a warning
  • Syncs commits, issues, PRs, reviews, comments, and more

Common questions

Yes, list repositories from different orgs in the format "org1/repo1,org2/repo2" or use "org/*" for all repos in an org.

Use multiple personal access tokens separated by commas to load balance API quota across tokens.

Security: Open only requires read access to your GitHub. We never write, modify, or delete your content. All data is encrypted in transit and at rest. SOC 2 Type II certified and GDPR compliant.

Ready to connect GitHub?

AI Training → Sources → GitHub