HighAI coding agentsBLWK-AI-007

An agent is configured to bypass all permission prompts

What this means

An agent told never to ask — bypassPermissions, Zed default: allow, Goose GOOSE_MODE: auto, OpenCode permission: allow, Qwen Code yolo, Cursor CLI unrestricted, Open Interpreter auto_run.

How to fix it

Turn off the setting named in this finding (for example Claude Code's defaultMode: “bypassPermissions”, Zed's tool_permissions default “allow”, Goose's GOOSE_MODE: auto, OpenCode's permission: “allow”, Qwen Code's approvalMode “yolo”, Cursor CLI approvalMode “unrestricted”), and only run an agent without approvals inside a throwaway container or VM. It removes every guardrail between the model and your machine.

Check it yourself

Run a scan and look for BLWK-AI-007 in the results:

bulwark scan
bulwark ai scan

Installed the standalone CLI package instead of the app? The command is bulwarkctl.

References

• MITRE ATT&CK T1059

Related checks

• Project-supplied agent hooks run shell commands automatically · BLWK-AI-002
• An agent permission allowlist permits arbitrary command execution · BLWK-AI-006
• Project MCP servers are set to auto-enable without prompting · BLWK-AI-008
• VS Code chat auto-approve (“YOLO mode”) is enabled · BLWK-AI-009
• VS Code Workspace Trust is disabled · BLWK-AI-010
• A VS Code task runs automatically when the folder is opened · BLWK-AI-011