> The problem required the user to set up a GitHub MCP server with credentials that allowed access to both public and private repos, to configure some LLM to ha

> The problem required the user to set up a GitHub MCP server with credentials that allowed access to both public and private repos, to configure some LLM to have access to that MCP server, and then to explicitly submit a request to that LLM that explicitly said to read and parse arbitrary issues (including the one created earlier) and then just blindly parse and process and perform whatever those issues said to do, and then blindly make a publicly-visible update to a public repo with the results of those operation(s). To be fair, with all the AI craze, this is exactly what lots of people are going to do without thinking twice. You might say "well they shouldn't, stupid". True. But that's what guardrails are for, because people often are stupid.