New: Drizz now tests AI agents in simulated companies, with a graded report on every run. See how it works →

TwinsGitHub

A GitHub twin for testing AI agents.

Test agents that open issues, review pull requests and merge code. Repositories your agent can push to, with no real repo at risk.

twin / githubprivate · seeded · resets in one call
GitHub twinRepos, issues, pull requests and reviews.
  • Repos and branches
  • Issues and pull requests
  • Reviews and checks
  • Labels and milestones
  • Branch rules
  • One-call reset
Meet the twin

A stateful copy of GitHub, built for testing agents.

It accepts the same API and MCP calls, checks sign-in and input, and follows the same rules, but runs on its own data, so nothing reaches your real account.

What your agent can do
  • Repositories, branches and commits
  • Issues and pull requests
  • Reviews, comments and checks
  • Labels and milestones
  • Permissions and branch rules
Seeded state

Every run starts from a company that already has history. We seed:

  • Repositories with history
  • Open pull requests with reviews
  • Issues linked to tickets
  • Teams and permissions
In staging

Use GitHub in staging, without using production.

1

Validate every action

The twin checks what the agent sends the way GitHub would: required fields, permissions, rate limits, malformed input. Mistakes fail here, not in your account.

2

Start from a realistic state

Describe the company and the situation in plain English. Drizz seeds the twin to match, and resets it to that state before every run.

3

Replay tasks across tools

The GitHub twin shares its people, customers and history with the other twins, so a task that touches three tools runs the way it would at work.

Point your agent at it

Swap the address. Your agent code stays the same.

Replace the live GitHub endpoint with the twin’s. Everything your agent already does keeps working, on data that is yours to break.

  • Same API and MCP calls
  • Sign-in and its failures
  • Real errors
  • Rate limits
  • Permissions and rules
  • Connected data
  • Reset in one call
{
  "mcpServers": {
    "github": {
      "type": "http",
      "url": "<twin MCP endpoint>",
      "headers": { "Authorization": "Bearer <token>" }
    }
  }
}

REST works on the same data. Both are reset with one call.

One company

Connected to the rest of the company.

A pull request links to the Jira or Linear ticket it closes and the Slack thread where it was reviewed.

17 of 20 runs passed. 3 broke a rule.

Run your agent on the GitHub twin.