GitHub statistics show an enormous contribution of AI in product metrics, but founders are worried about degraded quality. A few months ago, Archestra posted an issue with a $900 bounty for new "MCP Apps" support. Initially, legitimate contributors proposed plans and submitted attempts, but soon AI bots arrived, taking the issue to 253 comments and poisoning the conversation.
The problem escalated into an epidemic, with AI accounts flooding not just the issue but the entire repo. Real conversations from contributors were getting buried under a wall of noise, triggering notifications for every team member watching the repo. One team member spent half a day every week cleaning out untested PRs and closing hallucinated issues.
To combat the issue, Archestra is now blocking AI accounts from creating issues, opening PRs, or leaving comments unless they have previously committed to the main branch. The company has implemented an onboarding process with ethical AI rules and a CAPTCHA, followed by a GitHub Action that adds their handle to an external contributors file and pushes a commit to the main branch.
The goal is to make the repo a comfortable space for legitimate contributors, responsible AI users, newbies, and seasoned engineers. By prioritizing quality over quantity, Archestra aims to create a great piece of software where everyone can contribute without it being overrun by AI bots.
