The fastest, cheapest, most efficient open-source agent harness. Run more than 500 agents on a $10 VPS.
Rust agent-orchestrationai-agentsai-assistantdesktopllmlocal-firstmcppersonal-aiprivacyrustsecond-braintauri
fix(security): block env from launching command executors
Steven Enamakele11620c95891
diff --git a/crates/openhuman-core/src/security/policy/command_checks.rs b/crates/openhuman-core/src/security/policy/command_checks.rsindex 2a9ff9c531..afe81f58c6 100644--- a/crates/openhuman-core/src/security/policy/command_checks.rs+++ b/crates/openhuman-core/src/security/policy/command_checks.rs@@ -662,7 +662,12 @@ impl SecurityPolicy {// Validate arguments for the commandlet args: Vec<String> = words.map(|w| w.to_ascii_lowercase()).collect();- if is_command_executor(base_cmd) || !self.is_args_safe(base_cmd, &args) {+ let launches_executor =+ base_cmd == "env" && args.iter().any(|arg| is_command_executor(arg));+ if is_command_executor(base_cmd)+ || launches_executor+ || !self.is_args_safe(base_cmd, &args)+ {return false;}}