{"version":"1.0.127","anchor":"enhanced-sandbox-command-filtering","canonical_anchor":"enhanced-sandbox-command-filtering","heading":"Enhanced Sandbox Command Filtering","tier":null,"area":null,"scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/1.0.127\/e\/enhanced-sandbox-command-filtering","release_url":"https:\/\/changelogs.core-directive.com\/v\/1.0.127","markdown":"### Enhanced Sandbox Command Filtering\n\n**What:** Added support for substring-based command filtering in the sandbox configuration through the new `tengu_sandbox_disabled_commands` feature gate.\n\n**How it works:**\n\nThe sandbox now supports two methods for disabling commands:\n\n- Exact command matching (existing `unsandboxedCommands` in settings)\n- Substring matching (new `tengu_sandbox_disabled_commands` feature flag)\n\n**Details:**\n\n- Substring matching allows blocking commands that contain specific patterns anywhere in the command string\n- The new `tengu_sandbox_disabled_commands` feature flag provides `commands` and `substrings` arrays\n- Both methods are combined when evaluating whether a command should bypass sandboxing\n- **Evidence**: `Qg2()` function at line 379010 in v1.0.127 (replaces `qu2()` at line 383844 in v1.0.126)"}