{"version":"2.1.277","anchor":"new-shell-command-danger-pattern-detector","canonical_anchor":"new-shell-command-danger-pattern-detector","heading":"New shell-command danger-pattern detector","tier":"notice","area":"Bash Safety","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.277\/e\/new-shell-command-danger-pattern-detector","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.277","markdown":"### New shell-command danger-pattern detector\n\nA new detector flags shell commands containing risky constructs like command substitution before auto-approving them\n\n**What**\n\nClaude Code now parses a shell command's structure (using a tree-sitter-based parser) to detect constructs considered unsafe to auto-approve without asking the user, including:\n\n- brace and parameter expansions\n\n- `$'...'`\/`$\"...\"` style quoting\n\n- command substitution (`$(...)`) and backticks\n\n- process substitution (`<(...)` and `>(...)`)\n\n- arithmetic expansion (`$[...]`)\n\n- malformed or incomplete redirects\n\nIf any of these appear in a command, it's flagged as unsafe to auto-approve.\n\n**Why**\n\nThese shell constructs can let a command do something other than what it appears to do at a glance, such as running hidden commands or reading unexpected input. Flagging them prevents Claude Code from auto-approving a command that looks safe but isn't.\n\n- Area: Bash Safety\n- Tier: You'll notice\n- Useful: 3\/5\n- Signal: 2\/5"}