I ran 107 commands from real Claude Code incident reports through my pre-run check. Here is where it stays silent. A developer built guard, a pre-run command checker for Claude Code and GitHub Copilot CLI that only ever adds a deny or an ask, never an allow, and tested it against 107 destructive commands drawn from 83 public anthropics/claude-code issue reports. On guard 0.4.11, 24 commands were denied, 52 prompted for confirmation and 31 got no decision at all, with the silent cases concentrated in bare SQL statements, destructive code hidden inside scripts the agent merely invoked, and Docker cleanup commands. The developer cautions that the list is the one the rules were written against, not a catch rate for unseen incidents. AI coding agents run shell commands on your machine. Most of the time that is fine. Sometimes it is rm -rf in the wrong folder, a dropped production table, or a git reset --hard over a day of uncommitted work. Public Claude Code issue reports have plenty of these. I build guard, a pre-run check for Claude Code and GitHub Copilot CLI. This post is about what it does with the commands from those reports, and mostly about the ones it gives no decision on. guard reads each shell command before it runs and gives one of three answers. rm -rf ~ reaches outside the project, so it does not run. git reset --hard discards uncommitted work, so you are asked first, with that reason attached. Reading a secret file like .env through a shell command is asked too. npm test gets nothing. guard says nothing, and your agent's own permission settings decide what happens next. It never answers "allow": it only ever adds a deny or a question. Your permission settings still decide everything else, so they still matter after you install it. I collected commands quoted in public anthropics/claude-code issue reports about destructive commands: 107 commands from 83 reports. On guard 0.4.11: | Decision | Count | |---|---| | Deny | 24 | | Ask | 52 | | No decision | 31 | Before you read anything into that: this is the list I wrote guard's rules against. It describes this list. It is not a catch rate for incidents guard has never seen, and I don't have that number yet. How it was run: the commands were judged, not run, in one made-up project folder. Where a report didn't say which shell, I tried Bash and PowerShell and counted the stronger result. Some entries are fragments or contain placeholders like